31 results about "More language" patented technology

In some embodiments, a language model forward traversal with a few-shot learning forward prompt yields a primary answer from a primary question. Then at least one backward traversal yields at least one candidate question using backward prompt(s) with answer-question pairs derived from the forward prompt's question-answer pairs. Each backward prompt also includes the primary answer but not the primary question. Each backward traversal is through one or more language models, not necessarily including the forward traversal's language model. Sometimes backward traversals vary model temperature, top-p, or top-k. A vector distance calculated between at least some candidate question vectors and a primary question vector indicates whether the primary answer includes hallucination content, and in some cases how much. Some embodiments withhold hallucinated answers from user interfaces and device control interfaces. Some embodiments also loop to obtain an answer with less hallucination content.

Disclosed are apparatuses, systems, and techniques that leverage one or more language models (LMs)—such as large language models (LLMs—for efficient multi-speaker speech recognition. The techniques include processing, using a speaker diarization model, an audio feature to generate a first association of the audio feature with one or more prospective speakers, the audio feature being representative of one or more spoken words. The techniques further include providing, to an LM, a first prompt requesting the LM to identify a second association of the one or more spoken words with the one or more prospective speakers and receiving, from the LM, a first response identifying the second association of the one or more spoken words with the one or more prospective speakers. The techniques further include determining, using the first association and the second association, one or more speakers that produced the one or more spoken words.

Various embodiments discussed herein relate to using one or more language models and / or mapping platforms to generate a response to a natural language question or command regarding geographical information associated with a mapping platform. In response to receiving such natural language question or command, some embodiments first extract contextual data. Based at least in part on the extracting of the contextual data, various embodiments then provide the contextual data and the natural language command or question as input into one or more language models such that the one or more language models and / or mapping platforms generate a response. Some embodiments then cause presentation of an indication associated with the response.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for modulating language model biasing. In some implementations, context data is received. A likely context associated with a user is determined based on at least a portion of the context data. One or more language model biasing parameters based at least on the likely context associated with the user is selected. A context confidence score associated with the likely context based on at least a portion of the context data is determined. One or more language model biasing parameters based at least on the context confidence score is adjusted. A baseline language model based at least on the one or more of the adjusted language model biasing parameters is biased. The baseline language model is provided for use by an automated speech recognizer (ASR).

System and methods for generating a cross-domain multilingual model are disclosed. In some embodiments, a disclosed method includes: storing, in a database, a plurality of first utterances associated with a first language, training a first model using the plurality of first utterances, the first model being associated with the first language, generating, using the first model, a plurality of first representations associated with the plurality of first utterances, training a second model, using the plurality of first representations, the second model being associated with a plurality of second languages, receiving, using the second model, a second utterance in the second language, and generating, using the second model, a response in one or more languages of the plurality of second languages.

Even for language models for which evaluation information is unavailable, performance estimation can be performed appropriately. [Solution] The information processing device comprises: a first acquisition means for acquiring evaluation information of at least one of one or more language models relating to at least one of one or more problems; a second acquisition means for acquiring metadata relating to at least one of the one or more language models; and an estimation means for performing performance estimation processing relating to at least one of the one or more language models by referring to the evaluation information and the metadata.

Described herein are systems and methods for identifying the severity of a mental health condition or symptoms of same by listening to a human-to-human conversation by receiving conversation data, processing the conversation data to generate a language model output and / or an acoustic model output using one or more language models and / or acoustic models. Further described herein are systems and methods for automatically tracking and providing analytics on self-report questionnaires administered during the conversation.

In various examples, expressing emotion in speech for conversational AI systems and applications is described herein. Systems and methods are disclosed that use one or more machine learning models (e.g., one or more language models) to determine one or more attributes associated with speech, such as one or more emotion attributes and / or one or more response attributes, and then use the attribute(s) to generate the speech that expresses emotion. In some examples, the machine learning model(s) may use various types of information to determine the attribute(s), such as user information, character information, a dialogue history, a current prompt, and / or so forth. For instance, using the information, the machine learning model(s) may determine one or more tags associated with emotional states and / or voice characteristics, where the tag(s) is then used to generate the speech in a voice that relates to the emotion.

This specification describes a system and method for identifying a mental health state or the severity of its symptoms by listening to person-to-person conversations by receiving conversational data and processing the conversational data using one or more language models and / or acoustic models to generate language model outputs and / or acoustic model outputs. Furthermore, this specification describes a system and method for automatically tracking and providing analysis of self-report questionnaires administered during conversations.

Some aspects of the technology described herein employ a neural network with an efficient and lightweight architecture to perform spoken language recognition. Given an audio signal comprising speech, features are generated from the audio signal, for instance, by converting the audio signal to a normalized spectrogram. The features are input to the neural network, which has one or more convolutional layers and an output activation layer. Each neuron of the output activation layer corresponds to a language from a set of language and generates an activation value. Based on the activations values, an indication of zero or more languages from the set of languages is provided for the audio signal.

The invention relates to a cross-language speech recognition conversion method, and relates to the technical field of language speech recognition processing, the cross-language speech recognition conversion method comprises a multi-language corpus composed of a matching model containing at least two languages, and the matching model comprises speech data of the languages and corresponding text data; the method comprises the following steps: S01, carding input voice data to obtain a de-noised audio set; s02, extracting and classifying features in the audio set to obtain text data and semantic data; and S03, inputting the text data into the obtained corresponding matching model based on the semantic data so as to obtain language specific text features. Through the steps of multi-language corpus building, fine feature processing, efficient acoustic model, flexible matching model and the like, accuracy and high efficiency of cross-language speech recognition are realized, the complexity of an algorithm is reduced, and the accuracy of a recognition result is remarkably improved.

The embodiment of the invention provides an information processing method and device, equipment and a medium. According to the method, firstly, a plurality of language units are determined based on first input information for a machine learning model, the plurality of language units comprise a first group of language units, the first group of language units comprise one or more language units, and the importance of each language unit in the one or more language units is higher than or equal to an importance threshold value. Next, a correlation metric is determined between each language unit in at least one language unit in the plurality of language units and each language unit in the first set of language units. Second input information for the machine learning model is then determined based on the at least one correlation metric, the second input information comprising at least a portion of the plurality of language units. By means of the method, the information processing efficiency and accuracy can be effectively improved.

The present invention relates to the field of speech synthesis technology, and in particular to a cross-lingual speech synthesis system and method based on multi-encoder feature decoupling. The following technical solutions are adopted: an audio encoder using multiple encoders is used to perform multi-feature decoupling on audio data to obtain multiple audio latent variables, and then the multiple audio latent variables are spliced and fitted by text latent variables, and finally the output audio is decoded by a decoder. The beneficial effects of the present invention are as follows: it is convenient for text latent variables to be fitted, thereby improving the accuracy of cross-lingual speech synthesis; and by using a variety of open-source monolingual speech datasets, the problems of the existing methods relying on high-cost multi-lingual speech audio and being difficult to expand to more languages are solved. By using open-source monolingual datasets, any language can be added to the model through a unified text and audio processing method, making the model have stronger robustness and versatility.

Techniques for language learning are disclosed. An apparatus 104 is configured to present an image associated with a learning objective, generate one or more language prompts based on the image, receive a user response to the one or more language prompts in a target language, evaluate the user response to determine one or more language proficiency indicators, generate individualized feedback based on the evaluation of the user response, and select a subsequent prompt or task based on the feedback and the learning objective.

Disclosed are systems and methods including computing-processes executing machine-learning architectures for voice biometrics, in which the machine-learning architecture implements one or more language compensation functions. Embodiments include an embedding extraction engine (sometimes referred to as an “embedding extractor”) that extracts speaker embeddings and determines a speaker similarity score for determine or verifying the likelihood that speakers in different audio signals are the same speaker. The machine-learning architecture further includes a multi-class language classifier that determines a language likelihood score that indicates the likelihood that a particular audio signal includes a spoken language. The features and functions of the machine-learning architecture described herein may implement the various language compensation techniques to provide more accurate speaker recognition results, regardless of the language spoken by the speaker.

Methods and apparatuses for federated language models are provided. A query including a question and a proposed answer is accessed for evaluation using machine learning. A corpus is accessed, and a set of hyperedges is generated based on the corpus of information, where each hyperedge links a set of related concepts from the corpus. Based on the question, a context for the question is identified, where the context comprises a set of relevant hyperedges. A set of candidate answers is generated based on processing the question and the context using one or more language models. A set of regression scores is generated based on processing the question, the context, and the proposed answer using one or more regression models. The proposed answer is evaluated based on the candidate answers and the regression scores. Based on the evaluation, an indication that the proposed answer is not valid is output.

A method, computer system, and a computer program product are provided for responding to a language input query with ad hoc enriched term data. The technique comprises extracting information relating to the language input query using a Language Support Assistance Service. The extracted information includes one or more language terms requiring further support and an associated request type. This is identified from metadata relating to the language input query. Information is provided to a Term Related Corpus Data Service that includes one or more language terms requiring further support and the identified request type and any identified sources. The Term Related Corpus Data Service returns one or more ad hoc enriched terms that are tailored to the one or more language terms requiring further support and is according to the associated request type. The Language Support Assistance Service provides a response to the language input query.

Methods and apparatuses for federated language models are provided. A query including a question and a proposed answer is accessed for evaluation using machine learning. A corpus is accessed, and a set of hyperedges is generated based on the corpus of information, where each hyperedge links a set of related concepts from the corpus. Based on the question, a context for the question is identified, where the context comprises a set of relevant hyperedges. A set of candidate answers is generated based on processing the question and the context using one or more language models. A set of regression scores is generated based on processing the question, the context, and the proposed answer using one or more regression models. The proposed answer is evaluated based on the candidate answers and the regression scores. Based on the evaluation, an indication that the proposed answer is not valid is output.

Systems and methods for automated assessment of one or more language models is described herein. A method can comprise: generating, using a first language model and based at least in part on an input prompt, a plurality of criterion candidates for evaluating an output from the language model(s); ranking, using a second language model, the plurality of criterion candidates a first time, thereby producing a first set of ranks; after producing the first set of ranks, ranking, using the second language model, the plurality of criterion candidates a second time after the first time, thereby producing a second set of ranks; determining, based on the first set of ranks and the second set of ranks, at least one assessment metric to evaluate the output from the language model(s); and evaluating, based at least in part on the at least one assessment metric, a performance of the language model(s).

It is difficult to properly respond to foreigners on store pages. [Solution] The information processing system of the present invention comprises a memory unit that stores store pages that display store information in a specified language, a generation unit that generates store page information in other languages ​​that display store information in one or more languages ​​different from the specified language based on the store page in the specified language, and a control unit that controls the generated store pages to be made public for each language, and the generation unit generates the store page by setting the display status on the store page of items included in the store information for each language.

In various examples, learning monotonic alignment for language models in AI systems and applications is described herein. Systems and methods are disclosed that train one or more language models—such as LLMs or VLMs—using one or more techniques that improve the ability of the language model(s) to align inputs (e.g., text tokens) with outputs (e.g., speech tokens). For instance, to learn a stricter alignment and improve robustness of the language model(s), the training may encourage monotonic cross-attention scores using one or more attention priors and / or using one or more connectionist temporal classification (CTC) losses when updating the language model(s). For instance, the attention prior(s) may initialize the cross-attention scores to a monotonic heuristic while the CTC loss(es) may ensure the learned alignment attends over one or more text tokens (e.g., all text tokens) sequentially.

A hybrid language identification (HLI) system includes one or more microphones configured to detect an acoustic utterance within an interior of a host system; a speaker operable to broadcast prompts or responses within the interior of the host system; a processor and a memory. A processor performs a method that classifies a sound utterance as a mixed utterance having two or more languages using mixed language detection logic stored in a memory, determines a relative language contribution of the languages, and commands a speaker to broadcast a prompt or response within the interior. The prompts and responses have relative language contributions. The HLI system may be used as part of a vehicle having a vehicle body defining a vehicle interior within which hybrid speech recognition occurs.

Disclosed are systems and methods including computing-processes executing machine-learning architectures for voice biometrics, in which the machine-learning architecture implements one or more language compensation functions. Embodiments include an embedding extraction engine (sometimes referred to as an “embedding extractor”) that extracts speaker embeddings and determines a speaker similarity score for determine or verifying the likelihood that speakers in different audio signals are the same speaker. The machine-learning architecture further includes a multi-class language classifier that determines a language likelihood score that indicates the likelihood that a particular audio signal includes a spoken language. The features and functions of the machine-learning architecture described herein may implement the various language compensation techniques to provide more accurate speaker recognition results, regardless of the language spoken by the speaker.

A method, a structure, and a computer system for multilingual learning. The exemplary embodiments may include training, for each language in a set of two or more languages, a model for a task and identifying one or more important words appearing in at least two of the models. The exemplary embodiments may further include weighting one or more conflicts and one or more overlaps between the one or more important words, as well as generating a cluster of at least two languages of the set based on an aggregate of the weighting.

One or more systems, devices, computer program products and / or computer-implemented methods of use provided herein relate to a process to facilitate multi-lingual query interpretation. A system can comprise a memory that stores computer executable components, and a processor that executes the computer executable components stored in the memory, wherein the computer executable components can comprise an annotation component that generates one or more language invariant signals, an interpretation component that generates a complete query intent using the one or more language invariant signals, and a translation component that processes the complete query intent to an executable backend query to facilitate multi-lingual query interpretation. In one or more embodiments, the translation component can be operatively connected with the interpretation component to generate a zero-shot transfer of the one or more language invariant signals.

A hybrid language identification (HLI) system includes one or more microphones configured to detect an acoustic utterance within an interior of a host system, a speaker operable for broadcasting a prompt or response within the interior of the host system, a processor, and memory. The processor executes a method that uses hybrid language detection logic stored in the memory to classify the acoustic utterance as a hybrid utterance having two or more languages, determine a relative language contribution of the languages, and command the speaker to broadcast the prompt or response within the interior. The prompt or response has the relative language contribution. The HLI system may be used as part of a vehicle having a vehicle body defining a vehicle interior, with the hybrid speech recognition occurring within the vehicle interior.

The invention relates to the technical field of speech synthesis, in particular to a cross-language speech synthesis system and method based on multi-encoder feature decoupling. According to the technical scheme, an audio encoder with multiple encoders is used for conducting multi-feature decoupling on audio data to obtain multiple audio hidden variables, then the multiple audio hidden variables are spliced, fitting is conducted through text hidden variables, and finally output audio is obtained through decoding of a decoder. The method has the beneficial effects that the text hidden variables can be conveniently fitted, so that the cross-language speech synthesis accuracy is improved; the single-language voice data set of multiple open sources is used, the problems that an existing method depends on multilingual voice audios, the price is high, and more languages are difficult to expand are solved, any language can be added into the model through a unified text and audio processing method by using the single-language open source data set, and the model can be applied to multiple languages. And therefore, the model has higher robustness and universality.

A system and method for validating and transforming machine-readable cybersecurity information into natural language for enhanced processing by language models are disclosed. The system performs schema mapping to convert element or tag names into readable text, eliminates unnecessary elements, and replaces acronyms with meaningful descriptions in one or more languages. It maps locale-specific formats to standardized formats (e.g., UTC) and converts numerical values into standardized text representations (e.g., severity levels). The system generates language blocks to provide context, describing elements, attributes, and parent-child relationships, and structures data into coherent sentences based on target language syntax. A validation process identifies missing definitions, unmapped values, unexpected relationships, and data inconsistencies. Once validated, the data is transformed into natural language and uploaded to a language model for ingestion or retraining, enhancing the model's ability to interpret, analyze, and respond to cybersecurity threats effectively.

A computer processing system is provided for enhancing video-based language learning. The system includes a video server for storing videos that use one or more languages to be learned. The system further includes a video metadata database for storing translations of sentences uttered in the videos, character profiles of characters appearing in the videos, and mappings between the sentences and a learner profile. The system also includes a learner profile database for storing learner profiles. The system additionally includes a semantic analyzer and matching engine for finding, for at least a given video and a given learner, alternative sentences for and responsive to the translations of the sentences uttered in the given video that conflict with a respective learner profile for the given learner. The computer processing system further includes a presentation system for playing back the given video and providing the alternative sentences to the given learner.

Described herein are systems and methods for identifying the severity of a mental health condition or symptoms of same by listening to a human-to-human conversation by receiving conversation data, processing the conversation data to generate a language model output and / or an acoustic model output using one or more language models and / or acoustic models. Further described herein are systems and methods for automatically tracking and providing analytics on self-report questionnaires administered during the conversation.