Managing audio metadata

Abstract

Systems and techniques are described herein for labelling audio. For instance, a method for labelling audio is provided. The method may include obtaining audio metadata related to audio data, wherein the audio metadata includes labels descriptive of audio events of the audio data; obtaining user input indicative of a modification to the audio metadata; and updating the audio metadata based on the user input.

Description

Qualcomm Ref. No. 2500999WQ 1MANAGING AUDIO METADATATECHNICAL FIELD

[0001] The present disclosure generally relates to managing audio metadata. For example, aspects of the present disclosure include systems and techniques for managing audio metadata.BACKGROUND

[0002] Audio data may be associated with audio metadata. The audio metadata may describe one or more aspects of the audio data.SUMMARY

[0003] The following presents a simplified summary relating to one or more aspects disclosed herein. Thus, the following summary should not be considered an extensive overview relating to all contemplated aspects, nor should the following summary be considered to identify key or critical elements relating to all contemplated aspects or to delineate the scope associated with any particular aspect. Accordingly, the following summary presents certain concepts relating to one or more aspects relating to the mechanisms disclosed herein in a simplified form to precede the detailed description presented below.

[0004] Systems and techniques are described for labelling audio. According to at least one example, a method is provided for labelling audio. The method includes: obtaining audio metadata related to audio data, wherein the audio metadata includes labels descriptive of audio events of the audio data; obtaining user input indicative of a modification to the audio metadata; and updating the audio metadata based on the user input.

[0005] In another example, an apparatus for labelling audio is provided that includes one or more memories configured to store audio metadata related to audio data and one or more processors (e.g., configured in circuitry) coupled to the one or more memories. The one or more processors are configured to: obtain the audio metadata related to the audio data, wherein the audio metadata includes labels descriptive of audio events of the audio data; obtain user input indicative of a modification to the audio metadata; and update the audio metadata based on the user input.

[0006] In another example, a non-transitory computer-readable medium is provided that has stored thereon instructions that, when executed by one or more processors, cause the one or morePolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 2 processors to: obtain audio metadata related to audio data, wherein the audio metadata includes labels descriptive of audio events of the audio data; obtain user input indicative of a modification to the audio metadata; and update the audio metadata based on the user input.

[0007] In another example, an apparatus for labelling audio is provided. The apparatus includes: means for obtaining audio metadata related to audio data, wherein the audio metadata includes labels descriptive of audio events of the audio data; means for obtaining user input indicative of a modification to the audio metadata; and means for updating the audio metadata based on the user input.

[0008] In some aspects, one or more of the apparatuses described herein is, can be part of, or can include an extended reality device (e.g., a virtual reality (VR) device, an augmented reality (AR) device, or a mixed reality (MR) device), a vehicle (or a computing device, system, or component of a vehicle), a mobile device (e.g., a mobile telephone or so-called “smart phone”, a tablet computer, or other type of mobile device), a smart or connected device (e.g., an Internet- of-Things (loT) device), a wearable device, a personal computer, a laptop computer, a video server, a television (e.g., a network-connected television), a robotics device or system, or other device. In some aspects, each apparatus can include an image sensor (e.g., a camera) or multiple image sensors (e g., multiple cameras) for capturing one or more images. In some aspects, each apparatus can include one or more displays for displaying one or more images, notifications, and / or other displayable data. In some aspects, each apparatus can include one or more speakers, one or more light-emitting devices, and / or one or more microphones. In some aspects, each apparatus can include one or more sensors. In some cases, the one or more sensors can be used for determining a location of the apparatuses, a state of the apparatuses (e.g., a tracking state, an operating state, a temperature, a humidity level, and / or other state), and / or for other purposes.

[0009] This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used in isolation to determine the scope of the claimed subject matter. The subject matter should be understood by reference to appropriate portions of the entire specification of this patent, any or all drawings, and each claim.

[0010] The foregoing, together with other features and aspects, will become more apparent upon referring to the following specification, claims, and accompanying drawings.Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 3BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Illustrative examples of the present application are described in detail below with reference to the following figures:

[0012] FIG. 1 is a block diagram illustrating an example system through which a user may interact with a system regarding audio data, according to various aspects of the present disclosure;

[0013] FIG. 2A is a block diagram illustrating an example system for managing audio metadata, according to various aspects of the present disclosure;

[0014] FIG. 2B is a block diagram illustrating another example system for managing audio metadata, according to various aspects of the present disclosure;

[0015] FIG. 3A and FIG. 3B include a diagram illustrating an example interaction between a user and an audio-interaction system, according to various aspects of the present disclosure;

[0016] FIG. 4 is a flow diagram illustrating an example process for managing audio-data labels, in accordance with aspects of the present disclosure;

[0017] FIG. 5 is a block diagram illustrating an example of a deep learning neural network that can be used to perform various tasks, according to some aspects of the disclosed technology;

[0018] FIG. 6 is a block diagram illustrating an example of a convolutional neural network (CNN), according to various aspects of the present disclosure;

[0019] FIG. 7 is a block diagram of an example transformer in accordance with some aspects of the disclosure; and

[0020] FIG. 8 is a block diagram illustrating an example computing-device architecture of an example computing device which can implement the various techniques described herein.DETAILED DESCRIPTION

[0021] Certain aspects of this disclosure are provided below. Some of these aspects may be applied independently and some of them may be applied in combination as would be apparent to those of skill in the art. In the following description, for the purposes of explanation, specific details are set forth in order to provide a thorough understanding of aspects of the application. However, it will be apparent that various aspects may be practiced without these specific details. The figures and description are not intended to be restrictive.Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 4

[0022] The ensuing description provides example aspects only, and is not intended to limit the scope, applicability, or configuration of the disclosure. Rather, the ensuing description of the exemplary aspects will provide those skilled in the art with an enabling description for implementing an exemplary aspect. It should be understood that various changes may be made in the function and arrangement of elements without departing from the spirit and scope of the application as set forth in the appended claims.

[0023] The terms “exemplary” and / or “example” are used herein to mean “serving as an example, instance, or illustration.” Any aspect described herein as “exemplary” and / or “example” is not necessarily to be construed as preferred or advantageous over other aspects. Likewise, the term “aspects of the disclosure” does not require that all aspects of the disclosure include the discussed feature, advantage, or mode of operation.

[0024] As mentioned above, audio data may be associated with audio metadata. The audio metadata may describe one or more aspects of the audio data. For example, an audio labeler may receive audio data as an input. The audio labeler may label audio events in the audio data. For example, the audio labeler may generate audio metadata including labels of audio events and corresponding timestamps.

[0025] Systems, apparatuses, methods (also referred to as processes), and computer-readable media (collectively referred to herein as “systems and techniques”) are described herein for managing audio metadata. For example, the systems and techniques described herein may manage audio events of audio metadata based on user feedback. For instances, a user may interact with an audio question-answering system to manage audio events such as creating, updating, or deleting audio events in audio metadata. The systems and techniques may use audio metadata, a chat history, and user inputs to generate responses and to the user inputs and to update the audio metadata based on the user inputs.

[0026] The systems and techniques may perform real-time updates to audio metadata based on user feedback. The systems and techniques may provide enhanced accuracy and relevance of responses (e g., of an audio question-answering system), leading to improved user experience (e.g., in interacting with an audio question-answering system).

[0027] The systems and techniques may allow the audio question-answering system to provide accurate and contextually relevant responses to user queries. Additionally, the systems andPolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 5 techniques may efficiently handle user feedback to improve system performance (e.g., performance of an audio question-answering system).

[0028] The systems and techniques may dynamically update audio metadata based on user feedback (e.g., in real time), ensuring continuous improvement and accuracy. The systems and techniques may adapt audio metadata based on new information and user corrections, making the audio metadata more reliable. By providing more accurate and relevant answers, the systems and techniques may significantly improve user satisfaction and trust.

[0029] As an example, a user may submit a query to an audio question-answering system. The systems and techniques may generate an answer and an indication to update audio metadata (e.g., an update-required flag). Based on the indication to update the audio metadata, the systems and techniques may update the audio metadata based on the user input. The audio question-answering system may use the updated audio metadata for subsequent queries, ensuring improved accuracy and relevance.

[0030] Continuously learns from user interactions, enhancing the overall experience. For example, the systems and techniques may track preferences of a user over time.

[0031] The systems and techniques are applicable to various audio question-answering system that store modality-specific data and use the modality-specific data for question answering.

[0032] Various aspects of the application will be described with respect to the figures below.

[0033] FIG. 1 is a block diagram illustrating an example system 100 through which a user may interact with a system regarding audio data 104, according to various aspects of the present disclosure. For example, audio-interaction system 102 may obtain audio data 104. A user may provide user input 106 to audio-interaction system 102 regarding audio data 104. Audiointeraction system 102 may generate response 108 responsive to user input 106, based on audio data 104.

[0034] Audio data 104 may include a number of audio events. For example, audio data 104 may include sound recordings of various events, such as a person speaking, a dog barking, a bell ringing, etc. Audio data 104 may be according to any suitable format, such as Motion-Picture Experts Group (MPEG) Audio Layer III (MP3), waveform audio file format (WAV), etc.

[0035] A user may provide user input 106 to audio-interaction system 102 through any suitable format, for example, the user may interact with audio-interaction system 102 via a text chatPolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 6 system or through an audio chat system. For example, the user may enter user input 106 as a text query into a text-based chat application. As another example, the user may speak user input 106 and a microphone may record user input 106 and convert the spoken user input 106 into a text format. User input 106 may relate to audio events of audio data 104. For example, user input 106 may include a user input regarding the audio events, such as “does audio data 104 include a dog barking?” or “at what time of audio data 104 does a bell ring?”

[0036] Audio-interaction system 102 may generate response 108 based on audio data 104 and responsive to user input 106. Response 108 may have any suitable format. In some aspects, response 108 may have the same format as user input 106. For example, response 108 may be a text response displayed at a display of a text-based chat application. As another example, response 108 may be a vocalized response output by a speaker.

[0037] FIG. 2A is a block diagram illustrating an example system 200a for managing audio metadata, according to various aspects of the present disclosure. Audio-interaction system 102 of FIG. 1 may implement system 200a. For example, audio data 202 may be an example of audio data 104 of FIG. 1, user input 208 may be an example of user input 106 of FIG. 1, and response 216 may be an example of response 108 of FIG. 1.

[0038] Labeler 204 may generate audio metadata 206 based on audio data 202. Labeler 204 may be, or may include, a machine-learning model trained to label audio data. For example, labeler 204 may be, or may include, a classifier network trained to classify audio data into various pre-determined classes.

[0039] Labeler 204 may generate audio metadata 206 such that audio metadata 206 includes, for each audio event labeled by audio metadata 206, an identifier (e.g., a numerical identifier (ID)), a label, a start time, a stop time, a duration, and / or an occurrence indicator. In some aspects, audio metadata 206 may be generated, stored, and / or updated according to any suitable textbased format, such as a JavaScript Object Notation (JSON) format, an Extensible Markup Language (XML) format, a Comma Separated Values (CSV) format, or as a natural language string. Audio metadata 206 may be, or may include, strings such that a language model 214 and / or language model 220 may extract labels and timestamps therefrom. Additionally or alternatively, audio metadata 206 may be of a format that can be processed using regular expressions.Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 7

[0040] The identifier of an audio event may be used in the operation of system 200a to reference audio events. Using an identifier (e.g., instead of a label) may reduce a number of tokens to generate (e.g., by language model 220 in update instructions 224).

[0041] The label of an audio event may be based on the class into which labeler 204 classified the audio event. The label of an audio event may be descriptive of the audio event. For example, an audio event may be labeled “silence,” “church bell,” “bird chirping,” etc.

[0042] The start time of an audio event may indicate a time, relative to audio data 202 (e.g., t = 2.4 seconds after the beginning of audio data 202), or relative to an absolute date and time, of the beginning of the audio event. The stop time of an audio event may indicate a time of the end of the audio event. The duration of an audio event may be a time difference between when the audio event began and when the audio event ceased.

[0043] An audio event may recur multiple times in audio data. For example, a single audio file may include recordings of three separate instances of car horns honking. The occurrence indicator of audio metadata 206 may indicate which occurrence of an audio event an entry in audio metadata 206 refers to. For example, for an instance of audio data 202, a first entry of audio metadata 206 may have an ID of “1,” a label of “dog barking,” a start time of “5.0” (e.g., t = 5 seconds after the beginning of the instance of audio data 202), and an occurrence indicator of “1” (e.g., indicating this is the first entry in audio metadata 206 related to the label “dog barking”). A second entry of audio metadata 206, for the instance of audio data 202, may have an ID of “2,” a label of “dog barking,” a start time of “35.0,” and an occurrence indicator of “2” (e.g., indicating that this is the second entry in audio metadata 206 related to the label “dog barking”).

[0044] System 200a may manage audio metadata (e.g., audio metadata 206 and audio metadata 228). For example, system 200a may update or revise audio metadata 206 (or audio metadata 228) in response to user input 208. For instance, system 200a may obtain audio metadata 206 and user input 208. System 200a may modify audio metadata 206 based on user input 208 to generate audio metadata 228. Thereafter, system 200a may obtain a subsequent instance of user input 208. System 200a may modify audio metadata 228 based on the subsequent instance of user input 208. System 200a may sequentially update or revise audio metadata 228 based on any number of sequentially-received user inputs 208.

[0045] Language model 214 may generate response 216 and update indication 218 based on user input 208, prompt 212, and audio metadata 206 or audio metadata 228. Language model 214Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 8 may be, or may include, a large-language model (LLM). Language model 214 may be a machinelearning model trained to generate responses based on inputs, context data, and / or prompts.

[0046] Language model 214 may generate response 216 and update indication 218 based on audio metadata 206 or audio metadata 228. For example, language model 214 may generate response 216 and update indication 218 based on audio metadata 206 before system 200a has modified audio metadata 206 to generate audio metadata 228 or audio metadata 228 after system 200a has modified audio metadata 206 to generate audio metadata 228).

[0047] Language model 214 may generate response 216 as a response to a query or request of user input 208. Language model 214 may generate response 216 based on audio metadata 206 or audio metadata 228. For example, language model 214 may use audio metadata 206 or audio metadata 228 as the data on which to base response 216 to user input 208.

[0048] In some aspects, system 200a may have, store, maintain, and / or update history 210 based on user interactions with system 200a. History 210 may include instances of user inputs 208 and / or instances of responses 216. In some aspects, language model 214 may generate response 216 and / or update indication 218 based on history 210.

[0049] Prompt 212 may be, or may include, instructions for generating response 216 and update indication 218 based on user input 208 and audio metadata 206 or audio metadata 228 and optionally history 210. For example, prompt 212 may include instructions for language model 214 to use audio metadata 206 or audio metadata 228 as contextual information for responding to user input 208. Additionally or alternatively, prompt 212 may include instructions for language model 214 to use history 210 as additional contextual information for responding to user input 208.

[0050] Additionally, prompt 212 may include instructions for language model 214 to generate update indication 218. Prompt 212 may instruct language model 214 to determine whether to update audio metadata 206 or audio metadata 228 based on user input 208 and to output update indication 218 as an indication of whether to update audio metadata 206 or audio metadata 228. For example, update indication 218 may be a flag (e.g., a binary flag) indicative of a determination to update audio metadata 206 or audio metadata 228 based on user input 208.

[0051] Language model 214 may, or may not, be trained to determine whether updates to audio metadata 206 or audio metadata 228 are needed. For example, in some aspects, language modelPolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 9214 may be a generic LLM without specific training relative to the operations of system 200a. Prompt 212 may instruct language model 214 to generate update indication 218. In other aspects, language model 214 may be trained to generate update indication 218 based on user input 208, for example, through a supervised training process or a supervised fine-tuning process.

[0052] In cases in which update indication 218 indicates to update audio metadata 206 or audio metadata 228, language model 220 may generate update instructions 224. For example, system 200a may cause language model 220 to generate update instructions 224 based on user input 208, prompt 222, and audio metadata 206 or audio metadata 228. Language model 220 may be, or may include, an LLM. Language model 220 may be a machine-learning model trained to generate responses based on inputs, context data, and / or prompts. In some aspects, language model 220 may be the same LLM as language model 214. For example, language model 214 and language model 220 may be implemented using the same processing instructions and in the same hardware. Alternatively, language model 214 and language model 220 may be instances of the same LLM (e.g., implemented separately).

[0053] Language model 220 may generate update instructions 224 based on audio metadata 206 or audio metadata 228. For example, language model 220 may generate update instructions 224 based on audio metadata 206 before system 200a has modified audio metadata 206 to generate audio metadata 228 or audio metadata 228 after system 200a has modified audio metadata 206 to generate audio metadata 228).

[0054] Language model 220 may generate update instructions 224 based on user input 208 user input 208 and audio metadata 206 or audio metadata 228. For example, language model 220 may generate update instructions 224 as instructions for modifying audio metadata 206 or audio metadata 228 based on user input 208.

[0055] Prompt 222 may be, or may include, instructions for generating update instructions 224 based on user input 208 and audio metadata 206 or audio metadata 228. For example, prompt 222 may include instructions for language model 220 to use audio metadata 206 or audio metadata 228 as base information to be changed according to update instructions 224 to implement an intent expressed by user input 208.

[0056] Language model 220 may, or may not, be trained to generate update instructions 224. For example, in some aspects, language model 220 may be a generic LLM without specific training relative to the operations of system 200a. Prompt 222 may instruct language model 220Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 10 to update instructions 224. In other aspects, language model 220 may be trained to generate update instructions 224, for example, through a supervised training process or a supervised fine- tuning process.

[0057] Update instructions 224 may be, or may include, instructions, (e.g., logical instructions) for modifying audio metadata 206 or audio metadata 228 to generate audio metadata 228 (e.g., an updated instances of audio metadata 228).

[0058] Update instructions 224 (as generated by language model 220 based on prompt 222) may refer to entries in audio metadata 206 (or audio metadata 228) by identifier (ID) rather than by label. This may conserve computational resources by reducing a length of tokens of update instructions 224. Additionally, update instructions 224 (as generated by language model 220 based on prompt 222) may instruct one of three operations, for example, to create an entry, to update an entry, or to delete an entry. The instructions may be coded in update instructions 224 (as generated by language model 220 based on prompt 222) as “C” for create, “U” for update, or “D” for delete.

[0059] Updater 226 may update audio metadata 206 or audio metadata 228 based on update instructions 224. Updater 226 may follow logical instructions of update instructions 224 to modify audio metadata 206 or audio metadata 228 to generate the updated instances of audio metadata 228. In some aspects, updater 226 may modify audio metadata 206 to generate audio metadata 228 according to JSON formatting and / or protocols.

[0060] For example, audio metadata 206 may be, or may include:{ “Silence” : [ { “id”: “1”, “start_time”: 0.0, “end_time”: 2.1, “duration”: 2.1, “occurrence”: 1 } ],“Church bell”: [ { “id”: “2”, “start_time”: 2.4, “end_time”: 3.4, “duration”: 1.0, “occurrence”: 1 } ] }

[0061] Update instructions 224 may be, or may include:{ “operation”: ”C”, “id”: “3” , “data”; {“start_time”: 4.0, “end time”: 5.0, “duration”: 1.0, “occurrence”: 1 } }

[0062] For example, update instructions 224 may indicate an operation referred to as “C” (e.g., a “create” operation. Update instructions 224 may indicate the creation of a third (e.g., ID “3”)Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 11 entry in audio metadata 206, with a start time of “4.0,” an end time of “5.0,” a duration of “ 1.0,” and an occurrence indication of “1.”

[0063] Audio metadata 228 may be, or may include:{ “Silence” : [ { “id”: ”1” , “start_time”: 0.0, “end_time”: 2.1, “duration”: 2.1, “occurrence”: 1 } ],“Church bell”: [ {“id”: ”2”, “start_time”: 2.4, “end_time”: 3.4, “duration”: 1.0, “occurrence”: 1 } ] }“Bird chirping”: [ {“id”: ”3”, “start_time”: 4.0, “end_time”: 5.0, “duration”: 1.0, “occurrence”: 1 } ] }

[0064] FIG. 2B is a block diagram illustrating an example system 200b for managing audio metadata, according to various aspects of the present disclosure. System 200b is substantially similar to system 200a of FIG. 2A.

[0065] In system 200a of FIG. 2A, prompt 222 instruct language model 220 to generate update instructions 224, and updater 226 of system 200a generates audio metadata 228 based on update instructions 224. In contrast, in system 200b, prompt 232 instructs language model 220 to generate the updated instance of audio metadata 228.

[0066] FIG. 3A and FIG. 3B include a diagram illustrating an example interaction 300 between a user and an audio-interaction system (e.g., audio-interaction system 102 of FIG. 1), according to various aspects of the present disclosure. The audio-interaction system of FIG. 3A and FIG. 3B may implement, for example, system 200a of FIG. 2A or system 200b of FIG. 2B.

[0067] Audio metadata 302 is an example of audio metadata 206 of FIG. 2A or FIG. 2B. Initially, audio metadata 302 may be, or may include: audio meta data: {“Vehicle”: [ {“start_time”: 0.0,”end_time”: 1.8 “duration”: 1.8 “occurrence”: 1 }],“Knock”: [ {“start_time”: 1.0,”end_time”: 2.1 “duration”: 1.1 “occurrence”: 1 }],“Doorbell”: [“start_time”: 5.3, “end_time”: 7.0, “duration”: 1.7. “occurrence”: ! }],Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 12“Dog Barking”: [{“start-time”: 7.0, “end_time”: 7.7, “duration”: 0.7, “occurrence”: }, {“start_time”: 7.9, “end_time”: 9.0, “duration”: 1.1, “occurrence”: 2}, [“start time”: 9.2, “end_time”: 9.9, “duration”: 0.7: “occurrence”: 3}]}

[0068] User input 304 is an example of a first instances of user input 208 of FIG. 2A and FIG. 2B. For example, user input 304 may be, or may include, “What is happening in this audio?”

[0069] Response 306 may be an example of a first instances of response 216 of FIG. 2A and FIG. 2B. Response 306 may be, or may include: “In the audio, there is a vehicle sound for 1.8 seconds, a knock for 1.1 seconds, a doorbell ringing for 1.7 seconds, and three instances of dog barking for a total of 2.5 seconds.”

[0070] To generate response 306, system 200a of FIG. 2A or system 200b of FIG. 2B may provide user input 304, prompt 212, and audio metadata 302 to language model 214 and language model 214 may generate response 306 based on user input 304, prompt 212, and audio metadata 302.

[0071] Additionally, based on prompt 212, language model 214 may determine that user input 304 does not indicate a change to audio metadata 302. Accordingly, based on prompt 212, language model 214 may generate update indication 218, which may indicate no change to audio metadata 302. Based on update indication 218 not indicating a change to audio metadata 302, system 200a or system 200b may not cause language model 220 to generate update instructions 224 or audio metadata 228.

[0072] User input 308 is an example of a subsequent instance of user input 208. User input 308 may be or may include: “I don’t hear the knock sound. Can you please update this information?”

[0073] Based on user input 308, system 200a or system 200b may generate response 310 and update audio metadata 302 to become audio metadata 312. For example, system 200a or system 200b may provide user input 308, prompt 212, and audio metadata 302 to language model 214 and language model 214 may generate response 310 based on user input 308, prompt 212, and audio metadata 302.

[0074] Response 310 is an example of a subsequent instance of response 216. Response 310 may be or may include: “We appreciate your feedback; Based on your feedback, it seems therePolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 13 might be an issue with the knock sound in the audio. I will review the metadata to ensure accuracy.”

[0075] Additionally, based on prompt 212, language model 214 may determine that that user input 308 indicates a change to audio metadata 302. Accordingly, based on prompt 212, language model 214 may generate update indication 218, which may indicate a change to audio metadata 302. Based on update indication 218 indicating a change to audio metadata 302, system 200a may cause language model 220 to generate update instructions 224 and cause updater 226 to update audio metadata 228 to generate audio metadata 312. Alternatively, based on update indication 218 indicating a change to audio metadata 302, system 200b may cause language model 220 to update audio metadata 312 (e.g., based on prompt 232).

[0076] Audio metadata 312 is an example of audio metadata 228 of FIG. 2A or FIG. 2B. For example, audio metadata 312 may be, or may include: audio meta data: {“Vehicle”: [ {“start_time”: 0.0,”end_time”: 1.8 “duration”: 1.8 “occurrence”: 1 }],“Doorbell”: [“start_time”: 5.3, “end_time”: 7.0, “duration”: 1.7. “occurrence”: ! }],“Dog Barking”: [{“start-time”: 7.0, “end_time”: 7.7, “duration”: 0.7, “occurrence”: }, {“start_time”: 7.9, “end_time”: 9.0, “duration”: 1.1, “occurrence”: 2}, {“start_time”: 9.2, “end_time”: 9.9, “duration”: 0.7: “occurrence”: 3 }]}

[0077] Audio metadata 312 reflects changes to audio metadata 302 based on user input 308.

[0078] In some aspects, system 100 may modify audio metadata 302 to generate audio metadata 312 prior to outputting response 310. In some aspects, system 100 may output a response (e.g., a confirmation) based on modifying audio metadata 302.

[0079] User input 314 is an example of a subsequent instances of user input 208. User input 314 may be, or may include: “What are the audio events now?”

[0080] Response 316 is an example of a subsequent instance of response 216. Response 316 may be or may include: “The updated audio events are a vehicle sound for 1.8 seconds, a doorbell ringing for 1 .7 seconds, and three instances of dog barking for a total of 2.5 seconds.”Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 14

[0081] User input 314 is an example of a subsequent instances of user input 208. User input 314 may be or may include: “I also hear a hammer sound at 4.6 seconds. Can you please update the information?”

[0082] Based on user input 318, system 200a or system 200b may generate response 322 and update audio metadata 312 to become audio metadata 320. For example, system 200a or system 200b may provide user input 318, prompt 212, and audio metadata 312 to language model 214 and language model 214 may generate response 322 based on user input 318, prompt 212, and audio metadata 312.

[0083] Response 322 is an example of a subsequent instance of response 216. Response 322 may be or may include: “The updated audio events now include a vehicle sound for 1.8 seconds, a doorbell ringing for 1.7 seconds, three instances of dog barking for a total of 2.5 seconds, and a hammer sound at 4.6 seconds.”

[0084] Additionally, based on prompt 212, language model 214 may determine that that user input 308 indicates a change to audio metadata 312. Accordingly, based on prompt 212, language model 214 may generate update indication 218, which may indicate a change to audio metadata 312. Based on update indication 218 indicating a change to audio metadata 312, system 200a may cause language model 220 to generate update instructions 224 and cause updater 226 to update audio metadata 228 to generate audio metadata 320. Alternatively, based on update indication 218 indicating a change to audio metadata 312, system 200b may cause language model 220 to update audio metadata 312 (e.g., based on prompt 232).

[0085] Audio metadata 320 is an example of audio metadata 228 of FIG. 2A or FIG. 2B. For example, audio metadata 320 may be, or may include: audio meta data: {“Vehicle”: [ {“start time”: 0.0, ’’end time”: 1.8 “duration”: 1.8 “occurrence”: 1 }],“Hammer”: [“start_time”: 4.6, “end_time”: 5.4, “duration”: 0.8. “occurrence”: ! }],“Doorbell”: [“start_time”: 5.3, “end_time”: 7.0, “duration”: 1.7. “occurrence”: ! }],Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 15“Dog Barking”: [{“start-time”: 7.0, “end_time”: 7.7, “duration”: 0.7, “occurrence”: }, {“start_time”: 7.9, “end_time”: 9.0, “duration”: 1.1, “occurrence”: 2}, {“start_time”: 9.2, “end_time”: 9.9, “duration”: 0.7: “occurrence”: 3}]}

[0086] FIG. 4 is a flow diagram illustrating an example process 400 for managing audio-data labels, in accordance with aspects of the present disclosure. One or more operations of process 400 may be performed by a computing device (or apparatus) or a component (e.g., a chipset, codec, etc.) of the computing device. The computing device may be a mobile device (e.g., a mobile phone), a network-connected wearable such as a watch, an extended reality (XR) device such as a virtual reality (VR) device or augmented reality (AR) device, a vehicle or component or system of a vehicle, a desktop computing device, a tablet computing device, a server computer, a robotic device, and / or any other computing device with the resource capabilities to perform the one or more operations of process 400. The one or more operations of process 400 may be implemented as software components that are executed and run on one or more processors.

[0087] At block 402, a computing device (or one or more components thereof) may obtain audio metadata related to audio data, wherein the audio metadata includes labels descriptive of audio events of the audio data. For example, system 200a of FIG. 2A or system 200b of FIG. 2B may obtain audio metadata 206. Audio metadata 206 may be, or may include, labels descriptive of audio events.

[0088] In some aspects, the audio metadata further comprises timestamps associated with the labels. For example, audio metadata 206 may include timestamps associated with the labels and / or the audio events.

[0089] In some aspects, to obtain the audio metadata, the computing device (or one or more components thereof) may process the audio data using a classifier to generate the audio metadata. For example, labeler 204 may generate audio metadata 206 based on audio data 202.

[0090] In some aspects, the classifier may be, or may include, a machine-learning model trained to label audio inputs. For example, labeler 204 may be, or may include, a machine-learning model trained to label audio inputs (e.g., generate labels and / or time stamps for audio events).

[0091] In some aspects, the audio metadata is formatted according to: JavaScript Object Notation (JSON); Extensible Markup Language (XML); Comma Separated Values (CSV); or aPolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 16 natural language string. For example, audio metadata 206 may be formatted according to JSON, XML, CSV, or a natural string.

[0092] At block 404, the computing device (or one or more components thereof) may obtain user input indicative of a modification to the audio metadata. For example, system 200a may obtain user input 208, which may be indicative of a modification to audio metadata 206.

[0093] In some aspects, the computing device (or one or more components thereof) may process the audio metadata and the user input using a language model to determine whether to update the audio metadata. For example, language model 214 may proceed user input 208 to determine whether to update audio metadata 206.

[0094] In some aspects, to determine to update the audio metadata, the computing device (or one or more components thereof) may process, using the language model the audio metadata, the user input, a history of user inputs, and a prompt. For example, to determine whether to update audio metadata 206, language model 214 may process audio metadata 206, user input 208, optionally history 210, and prompt 212.

[0095] In some aspects, the prompt may include instructions for the language model to determine whether to update the audio metadata based on the user input. For example, prompt 212 may include instructions for language model 214 to determine whether to update audio metadata 206 based on user input 208.

[0096] In some aspects, the prompt may include instructions for the language model to output an indication of whether to update the audio metadata. For example, prompt 212 may include instructions for language model 214 to output update indication 218, which may be indicative of whether to update audio metadata 206.

[0097] In some aspects, the language model may be, or may include, a machine-learning model trained to generate output text based on input texts and prompts. For example, language model 214 may be a machine-learning model trained to generate output text based on input text.

[0098] At block 406, the computing device (or one or more components thereof) may update the audio metadata based on the user input. For example, system 200a may update audio metadata 206 based on user input 208.

[0099] In some aspects, to update the audio metadata, the computing device (or one or more components thereof) may process the user input, the audio metadata, and a prompt using aPolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 17 language model to generate updated audio metadata. For example, language model 220 may process user input 208 and audio metadata 206 based on prompt 232 to generate audio metadata 228, for example, as illustrated and described with regard to FIG. 2B.

[0100] In some aspects, the prompt may include instructions for the language model to modify the audio metadata based on the user input. For example, prompt 232 may be, or may include, instructions for language model 220 to modify audio metadata 206 based on user input 208.

[0101] In some aspects, the language model comprises a machine-learning model trained to generate output text based on input texts and prompts. For example, language model 220 may be, or may include, a machine-learning model trained to generate output text based on input texts and prompts.

[0102] In some aspects, process the audio metadata and the user input using a language model to determine whether to update the audio metadata, wherein the audio metadata is updated responsive to determining to update the audio metadata. For example, language model 214 may process audio metadata 206 and user input 208 to determine whether to update audio metadata 206 based on user input 208. Further, language model 220 may update audio metadata 206 based on update indication 218 (which may indicate whether language model 214 determined to update audio metadata 206).

[0103] In some aspects, to update the audio metadata, the computing device (or one or more components thereof) may: determine an update instruction for the audio metadata; and perform the update instruction relative to the audio metadata. For example, language model 220 may generate update instructions 224 and updater 226 may update audio metadata 206 to generate audio metadata 228, for example, as illustrated and described with regard to FIG. 2A.

[0104] In some aspects, the language model comprises a machine-learning model trained to generate output text based on input texts and prompts. For example, language model 220 may be, or may include, a machine-learning model trained to generate output text based on input texts and prompts.

[0105] In some aspects, to determine the update instruction for the audio metadata, the computing device (or one or more components thereof) may process the user input, the audio metadata, and a prompt using a language model to generate the update instruction. For example,Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 18 language model 220 may process user input 208, audio metadata 206 and prompt 222 to generate update instructions 224.

[0106] In some aspects, the prompt may include instructions for the language model to generate the update instruction. For example, prompt 222 may include instructions for language model 220 to generate update instructions 224 based on user input 208.

[0107] In some aspects, the prompt may include instructions for the language model to generate the update instruction according to a format of the audio metadata. For example, prompt 222 may instruct language model 220 to generate update instructions 224 according to a format of audio metadata 206.

[0108] In some aspects, the format may be JavaScript Object Notation (JSON); Extensible Markup Language (XML); Comma Separated Values (CSV); or a natural language string. For example, audio metadata 206 may be formatted according to JSON, XML, CSV, or a natural language string. Prompt 222 may instruct language model 220 to generate update instructions 224 to instruct updater 226 to generate audio metadata 228 according to the same format.

[0109] In some aspects, the audio metadata further comprises timestamps associated with the labels. For example, audio metadata 206 may include timestamps associated with the labels.

[0110] In some aspects, the prompt may include instructions for the language model to generate the update instruction such that the update instruction includes instructions to at least one of: add a new label to the audio metadata; modify an existing label of the audio data; or remove a label from the audio data. For example, prompt 222 may instruct language model 220 to generate update instructions 224 to instruct updater 226 to add a new label to audio metadata 206, modify an existing label of audio metadata 206, and / or remove a label from audio metadata 206.

[0111] In some aspects, the prompt may include instructions for the language model to generate the update instruction such that the update instruction includes instructions to at least one of: add a new label and a new associated timestamp to the audio metadata; modify an existing label of the audio data; modify an existing timestamp of the audio data; or remove a label and an associated timestamp from the audio data. For example, prompt 222 may instruct language model 220 to generate update instructions 224 to instruct updater 226 to add a new label and a new associated timestamp to audio metadata 206, modify an existing label of audio metadata 206,Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 19 modify an existing timestamp of audio data a label of 206, or remove a label and an associated timestamp from audio metadata 206.

[0112] In some aspects, to perform the update instruction relative to the audio metadata, computing device (or one or more components thereof) may modify the audio metadata based on the update instruction. For example, updater 226 may modify audio metadata 206 based on update instructions 224 to generate audio metadata 228.

[0113] In some aspects, process the audio metadata and the user input using a language model to determine whether to update the audio metadata, wherein the audio metadata is updated responsive to determining to update the audio metadata. For example, language model 214 may process audio metadata 206 and user input 208 to determine whether to update audio metadata 206 based on user input 208. Further, language model 220 may update audio metadata 206 based on update indication 218 (which may indicate whether language model 214 determined to update audio metadata 206).

[0114] In some aspects, the computing device (or one or more components thereof) may generate a response based on the user input. For example, language model 214 may generate response 216 based on user input 208 and audio metadata 206.

[0115] In some aspects, the computing device (or one or more components thereof) may include a microphone configured to capture the audio data. For example, system 200a may include a microphone to record audio data 202.

[0116] In some aspects, the computing device (or one or more components thereof) may include, or be coupled to, a user interface configured to obtain the user input. For example, system 200a may include a user interface, such as a microphone, keyboard, or a touch screen.

[0117] In some examples, as noted previously, the methods described herein (e.g., process 400 of FIG. 4, and / or other methods described herein) can be performed, in whole or in part, by a computing device or apparatus. In one example, one or more of the methods can be performed by system 100 of FIG. 1, system 200a of FIG. 2 A system 200b of FIG. 2B, or by another system or device. In another example, one or more of the methods (e.g., process 400, and / or other methods described herein) can be performed, in whole or in part, by the computing-device architecture 800 shown in FIG. 8. For instance, a computing device with the computing-device architecture 800 shown in FIG. 8 can include, or be included in, the components of the systemPolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 20100, system 200a and / or system 200b and can implement the operations of process 400, and / or other process described herein. In some cases, the computing device or apparatus can include various components, such as one or more input devices, one or more output devices, one or more processors, one or more microprocessors, one or more microcomputers, one or more cameras, one or more sensors, and / or other component(s) that are configured to carry out the steps of processes described herein. In some examples, the computing device can include a display, a network interface configured to communicate and / or receive the data, any combination thereof, and / or other component(s). The network interface can be configured to communicate and / or receive Internet Protocol (IP) based data or other type of data.

[0118] The components of the computing device can be implemented in circuitry. For example, the components can include and / or can be implemented using electronic circuits or other electronic hardware, which can include one or more programmable electronic circuits (e.g., microprocessors, graphics processing units (GPUs), digital signal processors (DSPs), central processing units (CPUs), and / or other suitable electronic circuits), and / or can include and / or be implemented using computer software, firmware, or any combination thereof, to perform the various operations described herein.

[0119] Process 400, and / or other process described herein are illustrated as logical flow diagrams, the operation of which represents a sequence of operations that can be implemented in hardware, computer instructions, or a combination thereof. In the context of computer instructions, the operations represent computer-executable instructions stored on one or more computer-readable storage media that, when executed by one or more processors, perform the recited operations. Generally, computer-executable instructions include routines, programs, objects, components, data structures, and the like that perform particular functions or implement particular data types. The order in which the operations are described is not intended to be construed as a limitation, and any number of the described operations can be combined in any order and / or in parallel to implement the processes.

[0120] Additionally, process 400, and / or other process described herein can be performed under the control of one or more computer systems configured with executable instructions and can be implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications) executing collectively on one or more processors, by hardware, or combinations thereof. As noted above, the code can be stored on a computer-readable or machine-Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 21 readable storage medium, for example, in the form of a computer program comprising a plurality of instructions executable by one or more processors. The computer-readable or machine- readable storage medium can be non-transitory.

[0121] As noted above, various aspects of the present disclosure can use machine-learning models or systems.

[0122] FIG. 5 is an illustrative example of a neural network 500 (e.g., a deep-learning neural network) that can be used to implement machine-learning based feature segmentation, implicit- neural-representation generation, rendering, classification, object detection, image recognition (e.g., face recognition, object recognition, scene recognition, etc.), feature extraction, authentication, gaze detection, gaze prediction, and / or automation. For example, neural network 500 may be an example of, or can implement, labeler 204 of FIG. 2A and FIG. 2B.

[0123] An input layer 502 includes input data. In one illustrative example, input layer 502 can include data representing audio data 202 of FIG. 2A and FIG. 2B. Neural network 500 includes multiple hidden layers, for example, hidden layers 506a, 506b, through 506n. The hidden layers 506a, 506b, through hidden layer 506n include “n” number of hidden layers, where “n” is an integer greater than or equal to one. The number of hidden layers can be made to include as many layers as needed for the given application. Neural network 500 further includes an output layer 504 that provides an output resulting from the processing performed by the hidden layers 506a, 506b, through 506n. In one illustrative example, output layer 504 can generate audio metadata 206 of FIG. 2 A and FIG. 2B.

[0124] Neural network 500 may be, or may include, a multi-layer neural network of interconnected nodes. Each node can represent a piece of information. Information associated with the nodes is shared among the different layers and each layer retains information as information is processed. In some cases, neural network 500 can include a feed-forward network, in which case there are no feedback connections where outputs of the network are fed back into itself. In some cases, neural network 500 can include a recurrent neural network, which can have loops that allow information to be carried across nodes while reading in input.

[0125] Information can be exchanged between nodes through node-to-node interconnections between the various layers. Nodes of input layer 502 can activate a set of nodes in the first hidden layer 506a. For example, as shown, each of the input nodes of input layer 502 is connected to each of the nodes of the first hidden layer 506a. The nodes of first hidden layer 506a canPolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 22 transform the information of each input node by applying activation functions to the input node information. The information derived from the transformation can then be passed to and can activate the nodes of the next hidden layer 506b, which can perform their own designated functions. Example functions include convolutional, up-sampling, data transformation, and / or any other suitable functions. The output of the hidden layer 506b can then activate nodes of the next hidden layer, and so on. The output of the last hidden layer 506n can activate one or more nodes of the output layer 504, at which an output is provided. In some cases, while nodes (e.g., node 508) in neural network 500 are shown as having multiple output lines, a node has a single output and all lines shown as being output from a node represent the same output value.

[0126] In some cases, each node or interconnection between nodes can have a weight that is a set of parameters derived from the training of neural network 500. Once neural network 500 is trained, it can be referred to as a trained neural network, which can be used to perform one or more operations. For example, an interconnection between nodes can represent a piece of information learned about the interconnected nodes. The interconnection can have a tunable numeric weight that can be tuned (e.g., based on a training dataset), allowing neural network 500 to be adaptive to inputs and able to learn as more and more data is processed.

[0127] Neural network 500 may be pre-trained to process the features from the data in the input layer 502 using the different hidden layers 506a, 506b, through 506n in order to provide the output through the output layer 504. In an example in which neural network 500 is used to identify features in images, neural network 500 can be trained using training data that includes both images and labels, as described above. For instance, training images can be input into the network, with each training image having a label indicating the features in the images (for the feature-segmentation machine-learning system) or a label indicating classes of an activity in each image. In one example using object classification for illustrative purposes, a training image can include an image of a number 2, in which case the label for the image can be [0 0 1 0 0 0 0 0 0 0].

[0128] In some cases, neural network 500 can adjust the weights of the nodes using a training process called backpropagation. As noted above, a backpropagation process can include a forward pass, a loss function, a backward pass, and a weight update. The forward pass, loss function, backward pass, and parameter update are performed for one training iteration. ThePolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 23 process can be repeated for a certain number of iterations for each set of training images until neural network 500 is trained well enough so that the weights of the layers are accurately tuned.

[0129] For the example of identifying objects in images, the forward pass can include passing a training image through neural network 500. The weights are initially randomized before neural network 500 is trained. As an illustrative example, an image can include an array of numbers representing the pixels of the image. Each number in the array can include a value from 0 to 255 describing the pixel intensity at that position in the array. In one example, the array can include a 28 x 28 x 3 array of numbers with 28 rows and 28 columns of pixels and 3 color components (such as red, green, and blue, or luma and two chroma components, or the like).

[0130] As noted above, for a first training iteration for neural network 500, the output will likely include values that do not give preference to any particular class due to the weights being randomly selected at initialization. For example, if the output is a vector with probabilities that the object includes different classes, the probability value for each of the different classes can be equal or at least very similar (e.g., for ten possible classes, each class can have a probability value of 0.1). With the initial weights, neural network 500 is unable to determine low-level features and thus cannot make an accurate determination of what the classification of the object might be. A loss function can be used to analyze error in the output. Any suitable loss function definition can be used, such as a cross-entropy loss. Another example of a loss function includes the mean squared error (MSE), defined as Etotai = 2 'A (target - output)2. The loss can be set to be equal to the value of Etotai.

[0131] The loss (or error) will be high for the first training images since the actual values will be much different than the predicted output. The goal of training is to minimize the amount of loss so that the predicted output is the same as the training label. Neural network 500 can perform a backward pass by determining which inputs (weights) most contributed to the loss of the network and can adjust the weights so that the loss decreases and is eventually minimized. A derivative of the loss with respect to the weights (denoted as d / dW, where W are the weights at a particular layer) can be computed to determine the weights that contributed most to the loss of the network. After the derivative is computed, a weight update can be performed by updating all the weights of the filters. For example, the weights can be updated so that they change in the opposite direction of the gradient. The weight update can be denoted as w = w1- t] dL / dW, where w denotes a weight, w, denotes the initial weight, and q denotes a learning rate. The learning ratePolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 24 can be set to any suitable value, with a high learning rate including larger weight updates and a lower value indicating smaller weight updates.

[0132] Neural network 500 can include any suitable deep network. One example includes a convolutional neural network (CNN), which includes an input layer and an output layer, with multiple hidden layers between the input and out layers. The hidden layers of a CNN include a series of convolutional, nonlinear, pooling (for downsampling), and fully connected layers. Neural network 500 can include any other deep network other than a CNN, such as an autoencoder, a deep belief nets (DBNs), a Recurrent Neural Networks (RNNs), among others.

[0133] FIG. 6 is an illustrative example of a convolutional neural network (CNN) 600. The input layer 602 of the CNN 600 includes data representing an image or frame. For example, the data can include an array of numbers representing the pixels of the image, with each number in the array including a value from 0 to 255 describing the pixel intensity at that position in the array. Using the previous example from above, the array can include a 28 x 28 x 3 array of numbers with 28 rows and 28 columns of pixels and 3 color components (e.g., red, green, and blue, or luma and two chroma components, or the like). The image can be passed through a convolutional hidden layer 604, an optional non-linear activation layer, a pooling hidden layer 606, and fully connected layer 608 (which fully connected layer 608 can be hidden) to get an output at the output layer 610. While only one of each hidden layer is shown in FIG. 6, one of ordinary skill will appreciate that multiple convolutional hidden layers, non-linear layers, pooling hidden layers, and / or fully connected layers can be included in the CNN 600. As previously described, the output can indicate a single class of an object or can include a probability of classes that best describe the object in the image.

[0134] The first layer of the CNN 600 can be the convolutional hidden layer 604. The convolutional hidden layer 604 can analyze image data of the input layer 602. Each node of the convolutional hidden layer 604 is connected to a region of nodes (pixels) of the input image called a receptive field. The convolutional hidden layer 604 can be considered as one or more filters (each filter corresponding to a different activation or feature map), with each convolutional iteration of a filter being a node or neuron of the convolutional hidden layer 604. For example, the region of the input image that a filter covers at each convolutional iteration would be the receptive field for the filter. In one illustrative example, if the input image includes a 28*28 array, and each filter (and corresponding receptive field) is a 5*5 array, then there will be 24*24Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 25 nodes in the convolutional hidden layer 604. Each connection between a node and a receptive field for that node learns a weight and, in some cases, an overall bias such that each node learns to analyze its particular local receptive field in the input image. Each node of the convolutional hidden layer 604 will have the same weights and bias (called a shared weight and a shared bias). For example, the filter has an array of weights (numbers) and the same depth as the input. A filter will have a depth of 3 for an image frame example (according to three color components of the input image). An illustrative example size of the filter array is 5 x 5 x 3, corresponding to a size of the receptive field of a node.

[0135] The convolutional nature of the convolutional hidden layer 604 is due to each node of the convolutional layer being applied to its corresponding receptive field. For example, a filter of the convolutional hidden layer 604 can begin in the top-left corner of the input image array and can convolve around the input image. As noted above, each convolutional iteration of the filter can be considered a node or neuron of the convolutional hidden layer 604. At each convolutional iteration, the values of the filter are multiplied with a corresponding number of the original pixel values of the image (e.g., the 5x5 filter array is multiplied by a 5x5 array of input pixel values at the top-left corner of the input image array). The multiplications from each convolutional iteration can be summed together to obtain a total sum for that iteration or node. The process is next continued at a next location in the input image according to the receptive field of a next node in the convolutional hidden layer 604. For example, a filter can be moved by a step amount (referred to as a stride) to the next receptive field. The stride can be set to 1 or any other suitable amount. For example, if the stride is set to 1, the filter will be moved to the right by 1 pixel at each convolutional iteration. Processing the filter at each unique location of the input volume produces a number representing the filter results for that location, resulting in a total sum value being determined for each node of the convolutional hidden layer 604.

[0136] The mapping from the input layer to the convolutional hidden layer 604 is referred to as an activation map (or feature map). The activation map includes a value for each node representing the filter results at each location of the input volume. The activation map can include an array that includes the various total sum values resulting from each iteration of the filter on the input volume. For example, the activation map will include a 24 x 24 array if a 5 x 5 filter is applied to each pixel (a stride of 1) of a 28 x 28 input image. The convolutional hidden layer 604 can include several activation maps in order to identify multiple features in an image. ThePolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 26 example shown in FIG. 6 includes three activation maps. Using three activation maps, the convolutional hidden layer 604 can detect three different kinds of features, with each feature being detectable across the entire image.

[0137] In some examples, a non-linear hidden layer can be applied after the convolutional hidden layer 604. The non-linear layer can be used to introduce non-linearity to a system that has been computing linear operations. One illustrative example of a non-linear layer is a rectified linear unit (ReLU) layer. A ReLU layer can apply the function f(x) = max(0, x) to all of the values in the input volume, which changes all the negative activations to 0. The ReLU can thus increase the non-linear properties of the CNN 600 without affecting the receptive fields of the convolutional hidden layer 604.

[0138] The pooling hidden layer 606 can be applied after the convolutional hidden layer 604 (and after the non-linear hidden layer when used). The pooling hidden layer 606 is used to simplify the information in the output from the convolutional hidden layer 604. For example, the pooling hidden layer 606 can take each activation map output from the convolutional hidden layer 604 and generates a condensed activation map (or feature map) using a pooling function. Max-pooling is one example of a function performed by a pooling hidden layer. Other forms of pooling functions be used by the pooling hidden layer 606, such as average pooling, L2-norm pooling, or other suitable pooling functions. A pooling function (e.g., a max-pooling filter, an L2-norm filter, or other suitable pooling filter) is applied to each activation map included in the convolutional hidden layer 604. In the example shown in FIG. 6, three pooling filters are used for the three activation maps in the convolutional hidden layer 604.

[0139] In some examples, max-pooling can be used by applying a max-pooling filter (e.g., having a size of 2x2) with a stride (e.g., equal to a dimension of the filter, such as a stride of 2) to an activation map output from the convolutional hidden layer 604. The output from a maxpooling filter includes the maximum number in every sub-region that the filter convolves around. Using a 2x2 filter as an example, each unit in the pooling layer can summarize a region of 2x2 nodes in the previous layer (with each node being a value in the activation map). For example, four values (nodes) in an activation map will be analyzed by a 2x2 max-pooling filter at each iteration of the filter, with the maximum value from the four values being output as the “max” value. If such a max-pooling filter is applied to an activation filter from the convolutional hiddenPolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 27 layer 604 having a dimension of 24x24 nodes, the output from the pooling hidden layer 606 will be an array of 12x12 nodes.

[0140] In some examples, an L2-norm pooling filter could also be used. The L2-norm pooling filter includes computing the square root of the sum of the squares of the values in the 2*2 region (or other suitable region) of an activation map (instead of computing the maximum values as is done in max-pooling) and using the computed values as an output.

[0141] The pooling function (e.g., max-pooling, L2-norm pooling, or other pooling function) determines whether a given feature is found anywhere in a region of the image and discards the exact positional information. This can be done without affecting results of the feature detection because, once a feature has been found, the exact location of the feature is not as important as its approximate location relative to other features. Max-pooling (as well as other pooling methods) offer the benefit that there are many fewer pooled features, thus reducing the number of parameters needed in later layers of the CNN 600.

[0142] The final layer of connections in the network is a fully-connected layer that connects every node from the pooling hidden layer 606 to every one of the output nodes in the output layer 610. Using the example above, the input layer includes 28 x 28 nodes encoding the pixel intensities of the input image, the convolutional hidden layer 604 includes 3 x24x24 hidden feature nodes based on application of a 5x5 local receptive field (for the filters) to three activation maps, and the pooling hidden layer 606 includes a layer of 3x 12x 12 hidden feature nodes based on application of max-pooling filter to 2x2 regions across each of the three feature maps. Extending this example, the output layer 610 can include ten output nodes. In such an example, every node of the 3x12x12 pooling hidden layer 606 is connected to every node of the output layer 610.

[0143] The fully connected layer 608 can obtain the output of the previous pooling hidden layer 606 (which should represent the activation maps of high-level features) and determines the features that most correlate to a particular class. For example, the fully connected layer 608 can determine the high-level features that most strongly correlate to a particular class and can include weights (nodes) for the high-level features. A product can be computed between the weights of the fully connected layer 608 and the pooling hidden layer 606 to obtain probabilities for the different classes. For example, if the CNN 600 is being used to predict that an object in an image is a person, high values will be present in the activation maps that represent high-level featuresPolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 28 of people (e.g., two legs are present, a face is present at the top of the object, two eyes are present at the top left and top right of the face, a nose is present in the middle of the face, a mouth is present at the bottom of the face, and / or other features common for a person).

[0144] In some examples, the output from the output layer 610 can include an M-dimensional vector (in the prior example, M=10). M indicates the number of classes that the CNN 600 has to choose from when classifying the object in the image. Other example outputs can also be provided. Each number in the M-dimensional vector can represent the probability the object is of a certain class. In one illustrative example, if a 10-dimensional output vector represents ten different classes of objects is [0 0 0.05 0.8 0 0.15 0 0 0 0], the vector indicates that there is a 5% probability that the image is the third class of object (e.g., a dog), an 80% probability that the image is the fourth class of object (e.g., a human), and a 15% probability that the image is the sixth class of object (e.g., a kangaroo). The probability for a class can be considered a confidence level that the object is part of that class.

[0145] FIG. 7 is a block diagram of an example transformer in accordance with some aspects of the disclosure. For example, transformer 700 may be an example of, or can implement, one or more of language model 214 and / or language model 220 of FIG. 2A and FIG. 2B.

[0146] In a convolutional neural network (CNN) model, the number of operations required to relate signals from two arbitrary input or output positions grows in the distance between positions, which makes learning dependencies at different distant positions challenging for a CNN model. A transformer 700 reduces the operations of learning dependencies by using an encoder 710 and a decoder 730 that implement an attention mechanism at different positions of a single sequence to compute a representation of that sequence. An attention function can be described as mapping a query and a set of key-value pairs to an output, where the query, keys, values, and output are all vectors. The output is computed as a weighted sum of the values, where the weight assigned to each value is computed by a compatibility function of the query with the corresponding key.

[0147] In one example of a transformer, the encoder 710 is composed of a stack of six identical layers and each layer has two sub-layers. The first sub-layer is a multi-head self-attention engine 712, and the second sub-layer is a fully-connected feed-forward network 714. A residual connection (not shown) connects around each of the sub-layers followed by normalization.Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 29

[0148] In this example transformer 700, the decoder 730 is also composed of a stack of six 6 identical layers. The decoder also includes a masked multi-head self-attention engine 732, a multi-head attention engine 734 over the output of the encoder 710, and a fully-connected feedforward network 726. Each layer includes a residual connection (not shown) around the layer, which is followed by layer normalization. The masked multi-head self-attention engine 732 is masked to prevent positions from attending to subsequent positions and ensures that the predictions at position i can depend only on the known outputs at positions less than i (e.g., autoregression).

[0149] In the transformer, the queries, keys, and values are linearly projected by a multi -head attention engine into learned linear projects, and then attention is performed in parallel on each of the learned linear projects, which are concatenated and then projected into final values.

[0150] The transformer also includes a positional encoder 740 to encode positions because the model does not contain recurrence and convolution and relative or absolute position of the tokens is needed. In the transformer 700, the positional encodings are added to the input embeddings at the bottom layer of the encoder 710 and the decoder 730. The positional encodings are summed with the embeddings because the positional encodings and embeddings have the same dimensions. A corresponding position decoder 750 is configured to decode the positions of the embeddings for the decoder 730.

[0151] In some aspects, the transformer 700 uses self-attention mechanisms to selectively weigh the importance of different parts of an input sequence during processing and allows the model to attend to different parts of the input sequence while generating the output. The input sequence is first embedded into vectors and then passed through multiple layers of self-attention and feed-forward networks. The transformer 700 can process input sequences of variable length, making it well-suited for natural language processing tasks where input lengths can vary greatly. Additionally, the self-attention mechanism allows the transformer 700 to capture long-range dependencies between words in the input sequence, which is difficult for RNNs and CNNs. The transformer with self-attention has achieved results in several natural language processing tasks that are beyond the capabilities of other neural networks and has become a popular choice for language and text applications. For example, the various large language models, such as a generative pretrained transformer (e g., ChatGPT, etc.) and other current models are types of transformer networks.Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 30

[0152] FIG. 8 illustrates an example computing-device architecture 800 of an example computing device which can implement the various techniques described herein. In some examples, the computing device can include a mobile device, a wearable device, an extended reality device (e.g., a virtual reality (VR) device, an augmented reality (AR) device, or a mixed reality (MR) device), a personal computer, a laptop computer, a video server, a vehicle (or computing device of a vehicle), or other device. For example, the computing-device architecture 800 may include, implement, or be included in any or all of system 100 of FIG. 1, system 200a of FIG. 2A, system 200b of FIG. 2B and / or other devices, modules, or systems described herein. Additionally or alternatively, computing-device architecture 800 may be configured to perform process 400, and / or other process described herein.

[0153] The components of computing-device architecture 800 are shown in electrical communication with each other using connection 812, such as a bus. The example computingdevice architecture 800 includes a processing unit (CPU or processor) 802 and computing device connection 812 that couples various computing device components including computing device memory 810, such as read only memory (ROM) 808 and random-access memory (RAM) 806, to processor 802.

[0154] Computing-device architecture 800 can include a cache of high-speed memory connected directly with, in close proximity to, or integrated as part of processor 802. Computingdevice architecture 800 can copy data from memory 810 and / or the storage device 814 to cache 804 for quick access by processor 802. In this way, the cache can provide a performance boost that avoids processor 802 delays while waiting for data. These and other modules can control or be configured to control processor 802 to perform various actions. Other computing device memory 810 may be available for use as well. Memory 810 can include multiple different types of memory with different performance characteristics. Processor 802 can include any general- purpose processor and a hardware or software service, such as service 1 816, service 2 818, and service 3 820 stored in storage device 814, configured to control processor 802 as well as a special-purpose processor where software instructions are incorporated into the processor design. Processor 802 may be a self-contained system, containing multiple cores or processors, a bus, memory controller, cache, etc. A multi-core processor may be symmetric or asymmetric.

[0155] To enable user interaction with the computing-device architecture 800, input device 822 can represent any number of input mechanisms, such as a microphone for speech, a touch-Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 31 sensitive screen for gesture or graphical input, keyboard, mouse, motion input, speech and so forth. Output device 824 can also be one or more of a number of output mechanisms known to those of skill in the art, such as a display, projector, television, speaker device, etc. In some instances, multimodal computing devices can enable a user to provide multiple types of input to communicate with computing-device architecture 800. Communication interface 826 can generally govern and manage the user input and computing device output. There is no restriction on operating on any particular hardware arrangement and therefore the basic features here may easily be substituted for improved hardware or firmware arrangements as they are developed.

[0156] Storage device 814 is a non-volatile memory and can be a hard disk or other types of computer readable media which can store data that are accessible by a computer, such as magnetic cassettes, flash memory cards, solid state memory devices, digital versatile discs (DVDs), cartridges, random-access memories (RAMs) 806, read only memory (ROM) 808, and hybrids thereof. Storage device 814 can include services 816, 818, and 820 for controlling processor 802. Other hardware or software modules are contemplated. Storage device 814 can be connected to the computing device connection 812. In one aspect, a hardware module that performs a particular function can include the software component stored in a computer-readable medium in connection with the necessary hardware components, such as processor 802, connection 812, output device 824, and so forth, to carry out the function.

[0157] The term “substantially,” in reference to a given parameter, property, or condition, may refer to a degree that one of ordinary skill in the art would understand that the given parameter, property, or condition is met with a small degree of variance, such as, for example, within acceptable manufacturing tolerances. By way of example, depending on the particular parameter, property, or condition that is substantially met, the parameter, property, or condition may be at least 90% met, at least 95% met, or even at least 99% met.

[0158] Aspects of the present disclosure are applicable to any suitable electronic device (such as security systems, smartphones, tablets, laptop computers, vehicles, drones, or other devices) including or coupled to one or more active depth sensing systems. While described below with respect to a device having or coupled to one light projector, aspects of the present disclosure are applicable to devices having any number of light projectors and are therefore not limited to specific devices.Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 32

[0159] The term “device” is not limited to one or a specific number of physical objects (such as one smartphone, one controller, one processing system and so on). As used herein, a device may be any electronic device with one or more parts that may implement at least some portions of this disclosure. While the below description and examples use the term “device” to describe various aspects of this disclosure, the term “device” is not limited to a specific configuration, type, or number of objects. Additionally, the term “system” is not limited to multiple components or specific aspects. For example, a system may be implemented on one or more printed circuit boards or other substrates and may have movable or static components. While the below description and examples use the term “system” to describe various aspects of this disclosure, the term “system” is not limited to a specific configuration, type, or number of objects.

[0160] Specific details are provided in the description above to provide a thorough understanding of the aspects and examples provided herein. However, it will be understood by one of ordinary skill in the art that the aspects may be practiced without these specific details. For clarity of explanation, in some instances the present technology may be presented as including individual functional blocks including functional blocks including devices, device components, steps or routines in a method embodied in software, or combinations of hardware and software. Additional components may be used other than those shown in the figures and / or described herein. For example, circuits, systems, networks, processes, and other components may be shown as components in block diagram form in order not to obscure the aspects in unnecessary detail. In other instances, well-known circuits, processes, algorithms, structures, and techniques may be shown without unnecessary detail in order to avoid obscuring the aspects.

[0161] Individual aspects may be described above as a process or method which is depicted as a flowchart, a flow diagram, a data flow diagram, a structure diagram, or a block diagram. Although a flowchart may describe the operations as a sequential process, many of the operations can be performed in parallel or concurrently. In addition, the order of the operations may be rearranged. A process is terminated when its operations are completed but could have additional steps not included in a figure. A process may correspond to a method, a function, a procedure, a subroutine, a subprogram, etc. When a process corresponds to a function, its termination can correspond to a return of the function to the calling function or the main function.

[0162] Processes and methods according to the above-described examples can be implemented using computer-executable instructions that are stored or otherwise available from computer-Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 33 readable media. Such instructions can include, for example, instructions and data which cause or otherwise configure a general-purpose computer, special purpose computer, or a processing device to perform a certain function or group of functions. Portions of computer resources used can be accessible over a network. The computer executable instructions may be, for example, binaries, intermediate format instructions such as assembly language, firmware, source code, etc.

[0163] The term “computer-readable medium” includes, but is not limited to, portable or nonportable storage devices, optical storage devices, and various other mediums capable of storing, containing, or carrying instruction(s) and / or data. A computer-readable medium may include a non-transitory medium in which data can be stored and that does not include carrier waves and / or transitory electronic signals propagating wirelessly or over wired connections. Examples of a non-transitory medium may include, but are not limited to, a magnetic disk or tape, optical storage media such as compact disk (CD) or digital versatile disk (DVD), flash memory, magnetic or optical disks, USB devices provided with non-volatile memory, networked storage devices, any suitable combination thereof, among others. A computer-readable medium may have stored thereon code and / or machine-executable instructions that may represent a procedure, a function, a subprogram, a program, a routine, a subroutine, a module, a software package, a class, or any combination of instructions, data structures, or program statements. A code segment may be coupled to another code segment or a hardware circuit by passing and / or receiving information, data, arguments, parameters, or memory contents. Information, arguments, parameters, data, etc. may be passed, forwarded, or transmitted via any suitable means including memory sharing, message passing, token passing, network transmission, or the like.

[0164] In some aspects the computer-readable storage devices, mediums, and memories can include a cable or wireless signal containing a bit stream and the like. However, when mentioned, non-transitory computer-readable storage media expressly exclude media such as energy, carrier signals, electromagnetic waves, and signals per se.

[0165] Devices implementing processes and methods according to these disclosures can include hardware, software, firmware, middleware, microcode, hardware description languages, or any combination thereof, and can take any of a variety of form factors. When implemented in software, firmware, middleware, or microcode, the program code or code segments to perform the necessary tasks (e.g., a computer-program product) may be stored in a computer-readable or machine-readable medium. A processor(s) may perform the necessary tasks. Typical examplesPolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 34 of form factors include laptops, smart phones, mobile phones, tablet devices or other small form factor personal computers, personal digital assistants, rackmount devices, standalone devices, and so on. Functionality described herein also can be embodied in peripherals or add-in cards. Such functionality can also be implemented on a circuit board among different chips or different processes executing in a single device, by way of further example.

[0166] The instructions, media for conveying such instructions, computing resources for executing them, and other structures for supporting such computing resources are example means for providing the functions described in the disclosure.

[0167] In the foregoing description, aspects of the application are described with reference to specific aspects thereof, but those skilled in the art will recognize that the application is not limited thereto. Thus, while illustrative aspects of the application have been described in detail herein, it is to be understood that the inventive concepts may be otherwise variously embodied and employed, and that the appended claims are intended to be construed to include such variations, except as limited by the prior art. Various features and aspects of the above-described application may be used individually or jointly. Further, aspects can be utilized in any number of environments and applications beyond those described herein without departing from the broader spirit and scope of the specification. The specification and drawings are, accordingly, to be regarded as illustrative rather than restrictive. For the purposes of illustration, methods were described in a particular order. It should be appreciated that in alternate aspects, the methods may be performed in a different order than that described.

[0168] One of ordinary skill will appreciate that the less than (“<“) and greater than (“>“) symbols or terminology used herein can be replaced with less than or equal to (“<”) and greater than or equal to (“>”) symbols, respectively, without departing from the scope of this description.

[0169] Where components are described as being “configured to” perform certain operations, such configuration can be accomplished, for example, by designing electronic circuits or other hardware to perform the operation, by programming programmable electronic circuits (e.g., microprocessors, or other suitable electronic circuits) to perform the operation, or any combination thereof.

[0170] The phrase “coupled to” refers to any component that is physically connected to another component either directly or indirectly, and / or any component that is in communication withPolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 35 another component (e.g., connected to the other component over a wired or wireless connection, and / or other suitable communication interface) either directly or indirectly.

[0171] Claim language or other language reciting “at least one of’ a set and / or “one or more” of a set indicates that one member of the set or multiple members of the set (in any combination) satisfy the claim. For example, claim language reciting “at least one of A and B” or “at least one of A or B” means A, B, or A and B. In another example, claim language reciting “at least one of A, B, and C” or “at least one of A, B, or C” means A, B, C, or A and B, or A and C, or B and C, A and B and C, or any duplicate information or data (e g., A and A, B and B, C and C, A and A and B, and so on), or any other ordering, duplication, or combination of A, B, and C. The language “at least one of’ a set and / or “one or more” of a set does not limit the set to the items listed in the set. For example, claim language reciting “at least one of A and B” or “at least one of A or B” may mean A, B, or A and B, and may additionally include items not listed in the set of A and B. The phrases “at least one” and “one or more” are used interchangeably herein.

[0172] Claim language or other language reciting “at least one processor configured to,” “at least one processor being configured to,” “one or more processors configured to,” “one or more processors being configured to,” or the like indicates that one processor or multiple processors (in any combination) can perform the associated operation(s). For example, claim language reciting “at least one processor configured to: X, Y, and Z” means a single processor can be used to perform operations X, Y, and Z; or that multiple processors are each tasked with a certain subset of operations X, Y, and Z such that together the multiple processors perform X, Y, and Z; or that a group of multiple processors work together to perform operations X, Y, and Z. In another example, claim language reciting “at least one processor configured to: X, Y, and Z” can mean that any single processor may only perform at least a subset of operations X, Y, and Z.

[0173] Where reference is made to one or more elements performing functions (e.g., steps of a method), one element may perform all functions, or more than one element may collectively perform the functions. When more than one element collectively performs the functions, each function need not be performed by each of those elements (e.g., different functions may be performed by different elements) and / or each function need not be performed in whole by only one element (e.g., different elements may perform different sub-functions of a function). Similarly, where reference is made to one or more elements configured to cause another element (e.g., an apparatus) to perform functions, one element may be configured toPolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 36 cause the other element to perform all functions, or more than one element may collectively be configured to cause the other element to perform the functions.

[0174] Where reference is made to an entity (e.g., any entity or device described herein) performing functions or being configured to perform functions (e.g., steps of a method), the entity may be configured to cause one or more elements (individually or collectively) to perform the functions. The one or more components of the entity may include at least one memory, at least one processor, at least one communication interface, another component configured to perform one or more (or all) of the functions, and / or any combination thereof. Where reference to the entity performing functions, the entity may be configured to cause one component to perform all functions, or to cause more than one component to collectively perform the functions. When the entity is configured to cause more than one component to collectively perform the functions, each function need not be performed by each of those components (e.g., different functions may be performed by different components) and / or each function need not be performed in whole by only one component (e.g., different components may perform different sub-functions of a function).

[0175] The various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the aspects disclosed herein may be implemented as electronic hardware, computer software, firmware, or combinations thereof. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

[0176] The techniques described herein may also be implemented in electronic hardware, computer software, firmware, or any combination thereof. Such techniques may be implemented in any of a variety of devices such as general-purposes computers, wireless communication device handsets, or integrated circuit devices having multiple uses including application in wireless communication device handsets and other devices. Any features described as modules or components may be implemented together in an integrated logic device or separately asPolsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 37 discrete but interoperable logic devices. If implemented in software, the techniques may be realized at least in part by a computer-readable data storage medium including program code including instructions that, when executed, performs one or more of the methods described above. The computer-readable data storage medium may form part of a computer program product, which may include packaging materials. The computer-readable medium may include memory or data storage media, such as random-access memory (RAM) such as synchronous dynamic random-access memory (SDRAM), read-only memory (ROM), non-volatile randomaccess memory (NVRAM), electrically erasable programmable read-only memory (EEPROM), flash memory, magnetic or optical data storage media, and the like. The techniques additionally, or alternatively, may be realized at least in part by a computer-readable communication medium that carries or communicates program code in the form of instructions or data structures and that can be accessed, read, and / or executed by a computer, such as propagated signals or waves.

[0177] The program code may be executed by a processor, which may include one or more processors, such as one or more digital signal processors (DSPs), general-purpose microprocessors, an application specific integrated circuits (ASICs), field programmable logic arrays (FPGAs), or other equivalent integrated or discrete logic circuitry. Such a processor may be configured to perform any of the techniques described in this disclosure. A general-purpose processor may be a microprocessor; but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, such as, a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. Accordingly, the term “processor,” as used herein may refer to any of the foregoing structure, any combination of the foregoing structure, or any other structure or apparatus suitable for implementation of the techniques described herein.

[0178] Illustrative aspects of the disclosure include:

[0179] Aspect 1 . An apparatus to label audio, the apparatus comprising: one or more memories configured to store audio metadata related to audio data; and one or more processors coupled to the one or more memories and configured to: obtain the audio metadata related to the audio data, wherein the audio metadata includes labels descriptive of audio events of the audio data; obtain user input indicative of a modification to the audio metadata; and update the audio metadata based on the user input.Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 38

[0180] Aspect 2. The apparatus of aspect 1, wherein, to obtain the audio metadata, the one or more processors is configured to process the audio data using a classifier to generate the audio metadata.

[0181] Aspect 3. The apparatus of aspect 2, wherein the classifier comprises a machine-learning model trained to label audio inputs.

[0182] Aspect 4. The apparatus of any one of aspects 1 to 3, wherein the one or more processors is configured to process the audio metadata and the user input using a language model to determine whether to update the audio metadata.

[0183] Aspect 5. The apparatus of aspect 4, wherein, to determine to update the audio metadata, the one or more processors is configured to process, using the language model the audio metadata, the user input, a history of user inputs, and a prompt.

[0184] Aspect 6. The apparatus of any one of aspects 4 or 5, wherein the prompt comprises instructions for the language model to determine whether to update the audio metadata based on the user input.

[0185] Aspect 7. The apparatus of any one of aspects 5 or 6, wherein the prompt comprises instructions for the language model to output an indication of whether to update the audio metadata.

[0186] Aspect 8. The apparatus of any one of aspects 6 or 7, wherein the language model comprises a machine-learning model trained to generate output text based on input texts and prompts.

[0187] Aspect 9. The apparatus of any one of aspects 1 to 8, wherein, to update the audio metadata, the one or more processors is configured to process the user input, the audio metadata, and a prompt using a language model to generate updated audio metadata.

[0188] Aspect 10. The apparatus of aspect 9, wherein the prompt comprises instructions for the language model to modify the audio metadata based on the user input.

[0189] Aspect 11. The apparatus of any one of aspects 9 or 10, wherein the one or more processors is configured to process the audio metadata and the user input using a language model to determine whether to update the audio metadata, wherein the audio metadata is updated responsive to determining to update the audio metadata.Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 39

[0190] Aspect 12. The apparatus of any one of aspects 9 to 11, wherein the language model comprises a machine-learning model trained to generate output text based on input texts and prompts.

[0191] Aspect 13. The apparatus of any one of aspects 1 to 12, wherein, to update the audio metadata, the one or more processors is configured to: determine an update instruction for the audio metadata; and perform the update instruction relative to the audio metadata.

[0192] Aspect 14. The apparatus of aspect 13, wherein, to determine the update instruction for the audio metadata, the one or more processors is configured to process the user input, the audio metadata, and a prompt using a language model to generate the update instruction.

[0193] Aspect 15. The apparatus of aspect 14, wherein the prompt comprises instructions for the language model to generate the update instruction.

[0194] Aspect 16. The apparatus of any one of aspects 14 or 15, wherein the prompt comprises instructions for the language model to generate the update instruction according to a format of the audio metadata.

[0195] Aspect 17. The apparatus of aspect 16, wherein the format comprises at least one of: JavaScript Object Notation (JSON); Extensible Markup Language (XML); Comma Separated Values (CSV); or a natural language string.

[0196] Aspect 18. The apparatus of any one of aspects 14 to 17, wherein the prompt comprises instructions for the language model to generate the update instruction such that the update instruction includes instructions to at least one of: add a new label to the audio metadata; modify an existing label of the audio data; or remove a label from the audio data.

[0197] Aspect 19. The apparatus of any one of aspects 14 to 18, wherein the audio metadata further comprises timestamps associated with the labels.

[0198] Aspect 20. The apparatus of aspect 19, wherein the prompt comprises instructions for the language model to generate the update instruction such that the update instruction includes instructions to at least one of: add a new label and a new associated timestamp to the audio metadata; modify an existing label of the audio data; modify an existing timestamp of the audio data; or remove a label and an associated timestamp from the audio data.Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 40

[0199] Aspect 21. The apparatus of any one of aspects 14 to 20, wherein the language model comprises a machine-learning model trained to generate output text based on input texts and prompts.

[0200] Aspect 22. The apparatus of any one of aspects 13 to 21, wherein, to perform the update instruction relative to the audio metadata, the one or more processors is configured to modify the audio metadata based on the update instruction.

[0201] Aspect 23. The apparatus of any one of aspects 13 to 22, wherein the one or more processors is configured to process the audio metadata and the user input using a language model to determine whether to update the audio metadata, wherein the update instruction is determined responsive to determining to update the audio metadata.

[0202] Aspect 24. The apparatus of any one of aspects 1 to 23, wherein the at least one processor is configured to generate a response based on the user input.

[0203] Aspect 25. The apparatus of any one of aspects 1 to 24, wherein the audio metadata further comprises timestamps related to the audio events.

[0204] Aspect 26. The apparatus of any one of aspects 1 to 25, wherein the audio metadata is formatted according to at least one of: JavaScript Object Notation (JSON); Extensible Markup Language (XML); Comma Separated Values (CSV); or a natural language string.

[0205] Aspect 27. The apparatus of any one of aspects 1 to 26, further comprising a microphone configured to capture the audio data.

[0206] Aspect 28. The apparatus of any one of aspects 1 to 27, further comprising a user interface configured to obtain the user input.

[0207] Aspect 29. A method for labelling audio, the method comprising: obtaining audio metadata related to audio data, wherein the audio metadata includes labels descriptive of audio events of the audio data; obtaining user input indicative of a modification to the audio metadata; and updating the audio metadata based on the user input.

[0208] Aspect 30. The method of aspect 29, wherein obtaining the audio metadata comprises processing the audio data using a classifier to generate the audio metadata.

[0209] Aspect 31. The method of aspect 30, wherein the classifier comprises a machinelearning model trained to label audio inputs.Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 41

[0210] Aspect 32. The method of any one of aspects 29 to 31, further comprising processing the audio metadata and the user input using a language model to determine whether to update the audio metadata.

[0211] Aspect 33. The method of aspect 32, wherein determining to update the audio metadata comprises processing, using the language model the audio metadata, the user input, a history of user inputs, and a prompt.

[0212] Aspect 34. The method of aspect 33, wherein the prompt comprises instructions for the language model to determine whether to update the audio metadata based on the user input.

[0213] Aspect 35. The method of any one of aspects 33 or 34, wherein the prompt comprises instructions for the language model to output an indication of whether to update the audio metadata.

[0214] Aspect 36. The method of any one of aspects 34 to 35, wherein the language model comprises a machine-learning model trained to generate output text based on input texts and prompts.

[0215] Aspect 37. The method of any one of aspects 29 to 36, wherein updating the audio metadata comprises processing the user input, the audio metadata, and a prompt using a language model to generate updated audio metadata.

[0216] Aspect 38. The method of aspect 37, wherein the prompt comprises instructions for the language model to modify the audio metadata based on the user input.

[0217] Aspect 39. The method of any one of aspects 37 or 38, further comprising processing the audio metadata and the user input using a language model to determine whether to update the audio metadata, wherein the audio metadata is updated responsive to determining to update the audio metadata.

[0218] Aspect 40. The method of any one of aspects 37 to 39, wherein the language model comprises a machine-learning model trained to generate output text based on input texts and prompts.

[0219] Aspect 41. The method of any one of aspects 29 to 40, wherein updating the audio metadata comprises: determining an update instruction for the audio metadata; and performing the update instruction relative to the audio metadata.Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 42

[0220] Aspect 42. The method of aspect 41, wherein determining the update instruction for the audio metadata comprises processing the user input, the audio metadata, and a prompt using a language model to generate the update instruction.

[0221] Aspect 43. The method of aspect 42, wherein the prompt comprises instructions for the language model to generate the update instruction.

[0222] Aspect 44. The method of any one of aspects 42 or 43, wherein the prompt comprises instructions for the language model to generate the update instruction according to a format of the audio metadata.

[0223] Aspect 45. The method of aspect 44, wherein the format comprises at least one of: JavaScript Object Notation (JSON); Extensible Markup Language (XML); Comma Separated Values (CSV); or a natural language string.

[0224] Aspect 46. The method of any one of aspects 42 to 45, wherein the prompt comprises instructions for the language model to generate the update instruction such that the update instruction includes instructions to at least one of: add a new label to the audio metadata; modify an existing label of the audio data; or remove a label from the audio data.

[0225] Aspect 47. The method of any one of aspects 42 to 46, wherein the audio metadata further comprises timestamps associated with the labels.

[0226] Aspect 48. The method of aspect 47, wherein the prompt comprises instructions for the language model to generate the update instruction such that the update instruction includes instructions to at least one of: add a new label and a new associated timestamp to the audio metadata; modify an existing label of the audio data; modify an existing timestamp of the audio data; or remove a label and an associated timestamp from the audio data.

[0227] Aspect 49. The method of any one of aspects 42 to 48, wherein the language model comprises a machine-learning model trained to generate output text based on input texts and prompts.

[0228] Aspect 50. The method of any one of aspects 41 to 49, wherein performing the update instruction relative to the audio metadata comprises modifying the audio metadata based on the update instruction.Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 43

[0229] Aspect 51. The method of any one of aspects 41 to 50, further comprising processing the audio metadata and the user input using a language model to determine whether to update the audio metadata, wherein the update instruction is determined responsive to determining to update the audio metadata.

[0230] Aspect 52. The method of any one of aspects 29 to 51, further comprising generating a response based on the user input.

[0231] Aspect 53. The method of any one of aspects 29 to 52, wherein the audio metadata further comprises timestamps related to the audio events.

[0232] Aspect 54. The method of any one of aspects 29 to 53, wherein the audio metadata is formatted according to at least one of: JavaScript Object Notation (JSON); Extensible Markup Language (XML); Comma Separated Values (CSV); or a natural language string.

[0233] Aspect 55. A non-transitory computer-readable storage medium having stored thereon instructions that, when executed by at least one processor, cause the at least one processor to perform operations according to any of aspects 29 to 54.

[0234] Aspect 56. An apparatus for providing virtual content for display, the apparatus comprising one or more means for perform operations according to any of aspects 29 to 54.Polsinelli Ref. No. 094922-863925

Claims

Qualcomm Ref. No. 2500999WQ 44CLAIMSWHAT IS CLAIMED IS:

1. An apparatus to label audio, the apparatus comprising: one or more memories configured to store audio metadata related to audio data; and one or more processors coupled to the one or more memories and configured to: obtain the audio metadata related to the audio data, wherein the audio metadata includes labels descriptive of audio events of the audio data; obtain user input indicative of a modification to the audio metadata; and update the audio metadata based on the user input.

2. The apparatus of claim 1, wherein, to obtain the audio metadata, the one or more processors is configured to process the audio data using a classifier to generate the audio metadata, and wherein the classifier comprises a machine-learning model trained to label audio inputs.

3. The apparatus of claim 1, wherein the one or more processors is configured to process the audio metadata and the user input using a language model to determine whether to update the audio metadata, and wherein the language model comprises a machine-learning model trained to generate output text based on input texts and prompts.

4. The apparatus of claim 3, wherein, to determine to update the audio metadata, the one or more processors is configured to process, using the language model the audio metadata, the user input, a history of user inputs, and a prompt.

5. The apparatus of claim 4, wherein the prompt comprises at least one of: instructions for the language model to determine whether to update the audio metadata based on the user input or instructions for the language model to output an indication of whether to update the audio metadata.

6. The apparatus of claim 1, wherein, to update the audio metadata, the one or more processors is configured to process the user input, the audio metadata, and a prompt using a language model to generate updated audio metadata.Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 457. The apparatus of claim 6, wherein the prompt comprises instructions for the language model to modify the audio metadata based on the user input.

8. The apparatus of claim 6, wherein the one or more processors is configured to process the audio metadata and the user input using the language model to determine whether to update the audio metadata, wherein the audio metadata is updated responsive to determining to update the audio metadata, and wherein the language model comprises a machine-learning model trained to generate output text based on input texts and prompts.

9. The apparatus of claim 1, wherein, to update the audio metadata, the one or more processors is configured to: determine an update instruction for the audio metadata; and perform the update instruction relative to the audio metadata.

10. The apparatus of claim 9, wherein, to determine the update instruction for the audio metadata, the one or more processors is configured to process the user input, the audio metadata, and a prompt using a language model to generate the update instruction, and wherein the language model comprises a machine-learning model trained to generate output text based on input texts and prompts.

11. The apparatus of claim 10, wherein the prompt comprises at least one of: instructions for the language model to generate the update instruction or instructions for the language model to generate the update instruction according to a format of the audio metadata.

12. The apparatus of claim 10, wherein the prompt comprises instructions for the language model to generate the update instruction such that the update instruction includes instructions to at least one of: add a new label to the audio metadata; modify an existing label of the audio data; or remove a label from the audio data.Polsinelli Ref. No. 094922-863925Qualcomm Ref. No. 2500999WQ 4613. The apparatus of claim 10, wherein the audio metadata further comprises timestamps associated with the labels.

14. The apparatus of claim 13, wherein the prompt comprises instructions for the language model to generate the update instruction such that the update instruction includes instructions to at least one of: add a new label and a new associated timestamp to the audio metadata; modify an existing label of the audio data; modify an existing timestamp of the audio data; or remove a label and an associated timestamp from the audio data.

15. The apparatus of claim 9, wherein, to perform the update instruction relative to the audio metadata, the one or more processors is configured to modify the audio metadata based on the update instruction.

16. The apparatus of claim 9, wherein the one or more processors is configured to process the audio metadata and the user input using a language model to determine whether to update the audio metadata, and wherein the update instruction is determined responsive to determining to update the audio metadata.

17. The apparatus of claim 1, wherein one or more processors is configured to generate a response based on the user input.

18. The apparatus of claim 1, wherein the audio metadata further comprises timestamps related to the audio events.

19. The apparatus of claim 1, further comprising a microphone configured to capture the audio data.

20. The apparatus of claim 1, further comprising a user interface configured to obtain the user input.Polsinelli Ref. No. 094922-863925

Images