A method and system for transcribing audio recordings

Abstract

The application discloses a recording transcription method and system, which is applied to a system comprising a wireless earphone, a charging bin and a mobile terminal. The method comprises the following steps: the wireless earphone and / or the charging bin start a recording mode, and collect first audio data and / or second audio data respectively; the wireless earphone and / or the charging bin send the collected audio data to the mobile terminal through a wireless transmission mode; the mobile terminal uploads the audio data to a cloud end for processing, receives a processing result returned by the cloud end, and displays the processing result on a terminal interface. Through the collaborative recording, seamless switching and intelligent cloud processing of the wireless earphone and the charging bin, the application realizes complete collection, continuous recording and instant information conversion of near-field and far-field dual-path audio, and significantly improves the flexibility, continuity and intelligent level of recording.

Description

Technical Field

[0001] This application relates to the field of headphone technology, and in particular to a recording transcription method and system. Background Technology

[0002] With the rapid development of wireless audio technology and the widespread adoption of mobile internet, wireless headphones (especially TWS headphones) have evolved from simple music playback devices into multifunctional smart wearable devices. Users increasingly demand recording functions from headphones for scenarios such as meeting minutes, study notes, and interviews. However, existing wireless headphones and charging case systems still have many limitations in practical applications, making it difficult to meet users' deeper needs for efficient and diverse recording capabilities.

[0003] First, the recording function is limited, resulting in incomplete sound field capture. Traditional wireless earbuds primarily focus on audio playback. While some products integrate microphones, their main purpose is limited to call noise reduction or simple voice command recognition, lacking independent, high-quality, long-term recording capabilities. Existing charging cases typically function only as power management units and lack audio capture functionality. This prevents users from simultaneously acquiring complete sound field information from both the wearer (near field) and the environment (far field). When users need to record their own speech and the voices of others around them, they often have to rely on a single device for time-sharing recording, leading to information gaps.

[0004] Secondly, data transmission and interaction methods are outdated, resulting in poor recording continuity. Most recording devices on the market, especially those with storage capabilities, rely on physical interfaces (such as wired USB connections) for data export. This wired transmission method is cumbersome, limiting ease of use, and lacks real-time performance, failing to meet the needs of immediate recording and processing. Furthermore, the interaction between the earphones and the charging case is usually limited to power transmission, lacking an effective data exchange channel. When the user places the earphones in the charging case, the earphones' recording function is often forced to stop, and the charging case cannot automatically resume recording, leading to the loss of crucial audio clips.

[0005] Finally, the lack of intelligent processing capabilities prevents the instant conversion of recordings into information. Most existing devices only offer local storage and simple playback of the raw audio, lacking intelligent processing capabilities such as real-time speech-to-text, text summarization, or translation. After recording, users still need to manually import the audio file into their computer or mobile phone and rely on third-party software for lengthy transcription and organization. This process significantly increases users' time costs, reduces work efficiency, and fails to meet the needs for instant information access in scenarios such as meeting minutes and study notes.

[0006] In summary, existing wireless earphone and charging case systems have significant shortcomings in terms of sound field acquisition integrity, wireless transmission convenience, recording continuity, and intelligent processing capabilities, and there is an urgent need for a solution that can overcome these limitations. Summary of the Invention

[0007] This application aims to address the technical deficiencies of existing wireless earphones and charging case systems in recording applications by providing a recording and transcription method and system. Through collaborative recording, seamless switching, and cloud-based intelligent processing between the wireless earphones and the charging case, it achieves complete acquisition, continuous recording, and real-time information conversion of near-field and far-field dual-channel audio, significantly improving the flexibility, continuity, and intelligence of recording.

[0008] In a first aspect, this application provides a recording transcription method, applied to a system including wireless earphones, a charging case, and a mobile terminal, the method comprising: The wireless earbuds and / or the charging case activate the recording mode to collect first audio data and / or second audio data respectively. The wireless earphones and / or the charging case will transmit the collected audio data to the mobile terminal wirelessly. The mobile terminal uploads the audio data to the cloud for processing, receives the processing results returned by the cloud, and displays them on the terminal interface.

[0009] Preferably, the cloud-based processing includes at least one of speech recognition to text, text summarization, and / or multilingual translation.

[0010] Preferably, the method further includes: During the simultaneous audio acquisition process of the wireless earbuds and the charging case, when it is detected that the wireless earbuds have been placed in the charging case; The charging case sends a stop recording command to the wireless earbuds, and the wireless earbuds stop recording and transmit the first audio data they have collected to the charging case for storage. The charging compartment continues to collect second audio data.

[0011] Preferably, the charging case detects whether the wireless earphones are placed inside the case via a reed switch or a Hall sensor.

[0012] Preferably, the charging case transmits the collected audio data to the mobile terminal wirelessly, including: The charging case packages the first audio data and the second audio data into a merged file or a separate file; The charging compartment transmits the merged file or the individual file to the mobile terminal wirelessly.

[0013] Preferably, the step of the wireless earphone transmitting the collected audio data to the mobile terminal via wireless transmission further includes: When the mobile terminal is out of wireless range or the wireless connection is unstable, the wireless earphones will send the first audio data to the charging case for caching. When the charging case detects that the wireless connection with the mobile terminal has been restored, it sends the first audio data to the mobile terminal.

[0014] Preferably, the wireless transmission method includes Bluetooth transmission and / or WiFi transmission; the step of the charging case sending the collected audio data to the mobile terminal further includes; When the second audio data collected by the charging case or the first audio data cached is high-quality audio data and / or long-duration audio data, WiFi transmission is used for transmission. When the second audio data collected by the charging case or the first audio data cached is low-quality audio data and / or short-duration audio data, Bluetooth transmission is used for transmission.

[0015] Preferably, the method further includes: after receiving the processing result returned by the cloud, the mobile terminal converts the processing result into voice and transmits it back to the wireless earphone via Bluetooth for playback.

[0016] Secondly, this application also provides an audio recording transcription system, the system comprising: Wireless earphones, used to collect near-field speech as the primary audio data; The charging case is used to charge the wireless earbuds and also to collect far-field or ambient voice as second audio data, and to cache the first audio data. The mobile terminal is used to receive audio data sent by the wireless earphones and / or the charging case, send it to the cloud for processing, and display the processing results fed back from the cloud. The cloud-based system is used for speech recognition, text conversion, and intelligent processing of audio data, and the processing results are returned to the mobile terminal.

[0017] Preferably, the mobile terminal is further configured to convert the processing result into voice information and send it to the wireless earphone for playback.

[0018] Beneficial effects: Compared with the prior art, the present invention has the following beneficial effects: Multi-directional audio acquisition enhances recording integrity: Audio can be acquired independently or simultaneously through both the wireless earbuds and the charging case, enabling dual-channel recording of near-field and far-field environmental speech. This meets the complete sound field requirements for various scenarios such as meetings, speeches, and interviews, avoiding information loss.

[0019] Collaborative recording and seamless switching ensure recording continuity: When the earbuds are placed in the charging case, the charging case automatically takes over the recording and receives the recorded data from the earbuds, enabling the recording task to continue continuously. This completely solves the problem of recording interruption caused by the earbuds being placed in the charging case, and avoids the loss of key audio segments.

[0020] Efficient wireless transmission, freeing you from wired constraints: Supports dual-mode wireless transmission via Bluetooth and WiFi, allowing users to export recording data anytime, anywhere without carrying a data cable; and adaptively selects the transmission method based on audio quality and duration (WiFi for high-quality / long-duration audio, Bluetooth for low-quality / short-duration audio), balancing transmission efficiency and power consumption.

[0021] Audio relay storage extends recording time and enhances data security: When the mobile terminal is unreachable, the earphones can back up the recording data to the charging case in real time via Bluetooth Low Energy. The charging case acts as a high-capacity storage relay, which not only solves the problems of small storage space and limited power consumption of the earphones, but also forms a double data protection to avoid recording loss.

[0022] Cloud-based intelligent processing enables instant conversion of recordings into information: It constructs an integrated closed loop of "recording - wireless transmission - cloud-based large-scale model processing - terminal display / voice feedback", automatically completing operations such as speech-to-text, text summarization, and multilingual translation. Users do not need to manually call third-party software, significantly improving work efficiency, especially suitable for scenarios such as meeting minutes and language learning.

[0023] Voice feedback enhances the interactive experience: Mobile terminals can synthesize text results processed in the cloud into speech and transmit them back to wireless headphones for playback, allowing users to obtain processed information in real time when it is inconvenient to look at the screen, thus improving the product's usability. Attached Figure Description

[0024] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0025] Figure 1 A schematic flowchart of a recording transcription method provided in this application embodiment; Figure 2 A schematic flowchart of the headphone recording transcription method provided in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of a recording transcription system provided in an embodiment of this application; Figure 4 A schematic flowchart of the headphone recording transcription method provided in the embodiments of this application; Figure 5 A schematic flowchart illustrating the charging case recording and transcription method provided in this application embodiment; Figure 6 A schematic flowchart illustrating the headphone charging case collaborative recording and transcription method provided in this application embodiment; The objectives, functional features, and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. The accompanying drawings have illustrated specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to specific embodiments. Detailed Implementation

[0026] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0027] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, components, features, and elements with the same names in different embodiments of this application may have the same meaning or different meanings, the specific meaning of which must be determined by its interpretation in that specific embodiment or further in conjunction with the context of that specific embodiment.

[0028] It should be understood that although the terms first, second, third, etc., may be used herein to describe various information, such information should not be limited to these terms. These terms are used only to distinguish information of the same type from one another. For example, without departing from the scope of this document, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if," as used herein, may be interpreted as "when," "when," or "in response to determination." Furthermore, as used herein, the singular forms "a," "an," and "the" are intended to also include the plural forms unless the context indicates otherwise. It should be further understood that the terms "comprising," "including," indicate the presence of the stated feature, step, operation, element, component, item, kind, and / or group, but do not exclude the presence, occurrence, or addition of one or more other features, steps, operations, elements, components, items, kinds, and / or groups. The terms "or," "and / or," "including at least one of the following," etc., as used in this application, may be interpreted as inclusive, or mean any one or any combination thereof. For example, "including at least one of the following: A, B, C" means "any one of the following: A; B; C; A and B; A and C; B and C; A and B and C." Similarly, "A, B, or C" or "A, B, and / or C" means "any one of the following: A; B; C; A and B; A and C; B and C; A and B and C." Exceptions to this definition only occur when the combination of elements, functions, steps, or operations is inherently mutually exclusive in some way.

[0029] It should be understood that although the steps in the flowcharts of this application's embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0030] Depending on the context, the words “if” or “suppose” as used here can be interpreted as “when” or “in response to determination” or “in response to detection.” Similarly, depending on the context, the phrases “if determination” or “if detection (of the stated condition or event)” can be interpreted as “when determination” or “in response to determination” or “when detection (of the stated condition or event)” or “in response to detection (of the stated condition or event).”

[0031] It should be noted that step designations such as S1 are used in this paper for the purpose of more clearly and concisely describing the corresponding content, and do not constitute a substantial restriction on the order.

[0032] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0033] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings and specific implementation methods in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0034] It should be noted that the "wireless earphones" involved in this invention include, but are not limited to, true wireless stereo (TWS) earphones and neckband wireless earphones; the "charging case" refers to a portable power box that provides charging, storage and additional functions for wireless earphones; the "mobile terminal" includes devices with wireless communication and display capabilities such as smartphones, tablets, and smartwatches; and the "cloud" refers to servers or server clusters that deploy large-scale models of speech recognition and natural language processing.

[0035] Example 1: refer to Figure 1 , Figure 2 As shown, the audio recording transcription method provided in this application includes: S1. The wireless earphones and / or the charging case start the recording mode and collect the first audio data and / or the second audio data respectively.

[0036] In the embodiments of this application, the user can start recording in any of the following ways: clicking the touch sensor on the wireless earbuds, pressing the physical button on the charging case, or clicking "Start Recording" through the APP interface on the mobile terminal. After starting, the user selects to execute the "earbud recording only", "charging case recording only", or "simultaneous recording" mode.

[0037] In this embodiment, in "earphone recording only" mode, the first microphone (MEMS microphone, signal-to-noise ratio ≥64dB) built into the wireless earphone collects the voice signal, which is then converted from analog to digital and compressed (e.g., using OPUS encoding, sampling rate 16kHz, bit rate 32kbps) by the earphone's internal audio codec (such as Cirrus Logic CS47L63) to generate the first audio data. The earphone temporarily stores the first audio data in an internal circular buffer (capacity 256KB).

[0038] In this embodiment, in the "charging case recording only" mode, the charging case's second microphone (dual microphone array, supporting beamforming) collects voice signals, and the charging case's main control chip (such as Realtek RTL8763E) writes the audio signals into the internal large-capacity storage unit (such as 16GB eMMC) to generate second audio data.

[0039] In this embodiment, in the "simultaneous recording" mode, the earphones and the charging case respectively collect their first audio data and second audio data in the manner described above, and achieve timestamp alignment (synchronization accuracy ±1ms) through Bluetooth clock synchronization. In a preferred embodiment, the first audio data collected by the earphones can be the user's near-field voice signal, and the second audio data collected by the charging case can be the ambient far-field voice.

[0040] In one embodiment, in the dual-microphone collaborative recording mode, when the wireless earphone is worn on the user's ear and the charging case is placed on a table (such as in a meeting scenario), the system controls the wireless earphone to start and collect the user's near-field voice (i.e., first audio data); at the same time, it controls the charging case to start its built-in microphone to collect ambient sound or far-field voice (i.e., second audio data).

[0041] In another embodiment, in single recording mode, if the earbuds are not in the charging case, only the charging case will record the environment; if the charging case is not nearby, only the earbuds will record while on the person.

[0042] S2. The wireless earphones and / or the charging case transmit the collected audio data to the mobile terminal wirelessly.

[0043] In the embodiments of this application, after the earphones and / or charging case collect audio data, the following methods are executed according to the transmission path preset by the user or automatically selected by the system: Headphone Direct Transmission: The headphones stream the first audio data to the mobile device in real time via classic Bluetooth (A2DP protocol, maximum bitrate 328kbps). This method is suitable for scenarios where real-time viewing of transcription results is required.

[0044] Direct transfer from charging case: The charging case can send the stored second audio file to the mobile terminal via the built-in WiFi module (2.4GHz, 802.11n) or classic Bluetooth.

[0045] Relay transmission: The earbuds first send data to the charging case via Bluetooth Low Energy (BLE 5.2), and then the charging case forwards it to the mobile terminal.

[0046] S3. The mobile terminal uploads the audio data to the cloud for processing, receives the processing result returned by the cloud, and displays it on the terminal interface.

[0047] In the embodiments of this application, after receiving audio data, the APP on the mobile terminal uploads the audio stream to the cloud server via Wi-Fi or cellular network. For example, the audio is uploaded in chunks (2 seconds each, approximately 32KB) to the cloud server's voice interface via the HTTPS / WebSocket protocol. After processing by the cloud, the result is returned, and the APP displays the transcribed text in real-time on the interface, while simultaneously showing a summary or translation. Users can click on any text paragraph to copy, share, or save it as a note.

[0048] In one embodiment of this application, after receiving audio data transmitted from a mobile terminal, the cloud processes the audio data, for example, converting the audio data into text information. In this embodiment, the cloud deploys a speech recognition engine (e.g., Whisper-large-v3) and a large language model (e.g., GPT-4o mini). The specific processing flow for processing the audio data is as follows: Speech-to-text conversion: After receiving the audio stream, the cloud uses an ASR model to convert the speech into timestamped text. For example, a 5-second meeting speech, "Project progress is normal, expected to launch next Friday," will be transcribed as [{"start":0.0, "end":5.0, "text":"Project progress is normal, expected to launch next Friday"}]. The system supports mixed Chinese and English recognition, with a word error rate of ≤5%.

[0049] Text Summary: Users can preset summary styles in the app (such as "one-sentence summary", "point-by-point summary", "to-do list"). The cloud sends the complete transcribed text into a large language model to generate a structured summary. For example, inputting a 30-minute meeting transcript will output "Resolution: Project postponed for two weeks; Responsible person: Zhang San; Next step: Submit test report by Friday".

[0050] Multilingual translation: Users select the source and target languages ​​(e.g., Chinese → Japanese). After transcribing the text, the cloud uses a translation model to translate sentence by sentence, maintaining timestamp correspondence. The translation results can be displayed in the form of the original text.

[0051] The three processing methods described above can be combined arbitrarily. When summarizing and translating are enabled simultaneously, the cloud first completes the transcription, then performs the summarization and translation separately, ultimately returning two independent result fields.

[0052] In one embodiment of this application, the user activates "Collaborative Recording Mode" via a mobile terminal APP. The wireless earbuds and charging case simultaneously begin collecting audio: the earbuds collect first audio data (e.g., near-field speech), and the charging case collects second audio data (far-field / ambient speech). During recording, the earbuds back up the collected audio data to the charging case via Bluetooth Low Energy.

[0053] When a user places the wireless earbuds into the charging case, the charging case detects the earbuds' insertion using a built-in reed switch (e.g., MKA-14103) or Hall sensor (e.g., DRV5032). In one embodiment, when the earbuds are inserted into the charging slot, a change in magnetic field strength causes the Hall sensor output level to flip. The charging case's microcontroller captures this signal via an external interrupt and samples it continuously for 50ms (5 times, 10ms interval) to confirm stability. Simultaneously, it checks for current (>10mA) at the charging contacts to double-confirm that the earbuds are in place. This detection response time is <20ms, meeting seamless switching requirements. The charging case's main controller immediately sends a stop recording command (command code 0x01) to the earbuds via Bluetooth Low Energy. Upon receiving the command, the earbuds stop microphone sampling and turn off the audio codec power. Simultaneously, the earbuds transmit the remaining first audio data in their internal buffer to the charging case via Bluetooth Low Energy. After receiving this data, the charging case continues to continuously collect second audio data, thus preventing recording interruptions.

[0054] In this embodiment, after the charging case finishes recording (either the user actively stops it or all earbuds have been removed), the data packaging steps are as follows: Packaged into a merged file: The first and second audio data are encoded into a stereo WAV file, with the left channel being near-field audio (from the headphones) and the right channel being far-field audio (from the charging case). The sampling rate is uniformly resampled to 16kHz, with 16-bit quantization. The original timestamp and synchronization offset are appended to the file header.

[0055] Package into separate files: Save as two MP3 (or WAV) files with filenames in the format "YYYYMMDD_HHMMSS_ear.mp3" and "YYYYMMDD_HHMMSS_env.mp3", and generate a JSON metadata file to record the correspondence between the two audio streams.

[0056] After packaging, the charging case sends the files to the mobile terminal based on network conditions (e.g., WiFi is prioritized if available, otherwise Bluetooth is used). Upon receiving the files, the mobile terminal can send them to the cloud for processing, either individually or in combination. In this embodiment, the charging case can also select a suitable wireless transmission method based on the quality and length of the audio data. For example, when the second audio data collected by the charging case or the first audio data cached is high-quality audio data (e.g., sampling rate ≥ 44.1kHz or bit rate ≥ 192kbps, or marked as "lossless mode" in the user's app) and / or long-duration audio data (e.g., single recording duration ≥ 10 minutes, or file size ≥ 50MB), during actual transmission, the charging case first checks whether it is connected to an available WiFi network (home / office WiFi or mobile hotspot). If already connected, upload will be made via HTTP POST. If not connected but the phone's hotspot is enabled, the charging case can quickly connect via Bluetooth assisted network pairing (sending the hotspot SSID and password). When the second audio data collected by the charging case or the first audio data cached is low-quality audio data (e.g., sampling rate ≤ 16kHz and bit rate ≤ 64kbps) and / or short-duration audio data (single recording < 10 minutes and file size < 50MB), Bluetooth transmission will be used if both conditions are met or WiFi is unavailable. If the network status changes during transmission (e.g., WiFi disconnection), the charging case supports resume transmission from where it left off: the already transmitted portion of the record offset will be resumed from the point of interruption after switching to Bluetooth. Users can also manually specify the transmission method or disable the adaptive function in the APP.

[0057] In one embodiment of this application, in "earphone recording only" mode, when the mobile terminal is not within the wireless connection range of the earphones or the wireless connection with the earphones is unstable, the first audio data collected is sent to the charging case for buffering. In one embodiment, when the wireless earphones detect a loss of Bluetooth connection with the mobile terminal or a signal strength below -80dBm (i.e., "not within wireless connection range or unstable connection"), the earphones automatically switch transmission strategies: every 30 seconds of audio collected by the earphones (approximately 960KB PCM, compressed to approximately 120KB in OPUS format), the data is sent to the charging case for buffering via Bluetooth Low Energy. The charging case stores the received first audio data in a dedicated relay buffer (e.g., a 4GB circular storage area in eMMC). A reception timestamp and original serial number are included during storage.

[0058] When the charging case detects a restored wireless connection with the mobile terminal, it sends the cached first audio data to the mobile terminal. In one embodiment, the charging case simultaneously and periodically scans for Bluetooth broadcasts or WiFi beacons on the mobile terminal. When the MAC address of the mobile terminal is detected and the signal strength is greater than -60dBm, the charging case actively initiates a connection. After the connection is established, the charging case uploads all cached audio files to the mobile terminal in chronological order via WiFi or Bluetooth. During the upload process, newly generated audio from the earbuds can continue to be stored in the charging case, achieving simultaneous storage and upload. If the charging case has less than 10% of its storage space remaining, it automatically deletes the earliest batch of successfully uploaded files or notifies the earbuds to reduce the bitrate (e.g., from 32kbps to 16kbps).

[0059] In one embodiment of this application, after the mobile terminal receives the processing result (e.g., transcribed text, summary, or translated text) returned from the cloud, the APP calls a local text-to-speech engine (e.g., Android TTS, iOS AVSpeechSynthesizer) or a cloud-based TTS API to synthesize the text into natural speech. The user can select different speakers (male / female voices), speaking speeds, and languages.

[0060] The synthesized speech is transmitted back to the wireless earphones in the following ways: Real-time transmission mode: When the cloud returns the transcription results sentence by sentence in a streaming manner, the APP immediately synthesizes speech upon receiving each complete sentence of text and sends it to the headphones in real time via a Bluetooth SCO link (connection-oriented synchronous link), where the headphones' speakers play the audio. In this mode, users can "dictate" the transcribed content without looking at their phone screen.

[0061] Post-recording playback mode: After all recordings have been processed, users can view the results on the app. Clicking the "play" button next to any paragraph (such as a meeting summary) will cause the app to synthesize the text into speech and send it to the headphones via the Bluetooth A2DP music channel. The headphones support single-click on the touchpad to pause / resume and double-click to skip to the next item.

[0062] During the transmission process, if the headset microphone detects the user speaking (voice activity detection), the app automatically lowers the TTS volume or pauses playback, resuming playback after the user stops speaking, thus achieving intelligent interruption.

[0063] Example 2: refer to Figure 3 As shown in the embodiments of this application, the recording and transcription system includes: wireless earphones, charging case, mobile terminal and cloud.

[0064] In this embodiment, the wireless earphones incorporate a first microphone (e.g., Knowles SPH1642, signal-to-noise ratio 64dB), an audio codec (e.g., Hengxuan BES2600), a Bluetooth 5.3 module (e.g., supporting classic Bluetooth and Bluetooth Low Energy), a touch sensor (e.g., TTP223), and a 30mAh lithium polymer battery. The earphones are responsible for collecting near-field voice data as the first audio data and can communicate with the charging case or mobile terminal via Bluetooth. The earphones have an internal temporary audio buffer (e.g., 256KB SRAM), which can store up to 15 seconds of audio when the Bluetooth connection is interrupted.

[0065] In this embodiment, the charging case includes: a second microphone (e.g., dual MEMS array, 15mm spacing), a main control chip (e.g., Realtek RTL8763E, integrating Bluetooth and WiFi), a large-capacity storage unit (e.g., 16GB eMMC), a Hall sensor (e.g., MT1401A), a 500mAh lithium-ion battery, physical buttons (e.g., recording start / stop, pairing, WiFi switch), and RGB status indicator lights. The charging case is used to charge the wireless earbuds, while simultaneously collecting far-field or ambient voice data as second audio data and buffering the first audio data sent from the wireless earbuds. The charging case can also transmit audio data to a mobile terminal at high speed via WiFi or Bluetooth.

[0066] In the embodiments of this application, the mobile terminal is a smartphone (iOS / Android) with a "Smart Recording" APP installed. This APP is responsible for receiving audio data sent from the wireless earbuds and / or charging case, uploading the audio data to the cloud for processing, and receiving the processing results returned from the cloud and displaying them on the interface. The APP also provides functions such as device management, recording mode selection, transmission strategy configuration, and history viewing.

[0067] In the embodiments of this application, the cloud deployment is carried out on GPU instances of Alibaba Cloud or AWS, running large models or interfaces such as the Whisper-large-v3 ASR model, the Tongyi Qianwen / BERT summary model, and the Google Translate API. The cloud performs speech recognition, text conversion, and intelligent processing on the audio data uploaded by the mobile terminal, and returns the processing results to the mobile terminal. Users can choose private deployment, deploying the cloud service on their own servers to protect privacy.

[0068] In embodiments of this application, the mobile terminal also has the function of converting the processing result into speech and sending it to the wireless headphones for playback. Specifically, after receiving the text result returned from the cloud, the APP displays a "Voice Broadcast" button on the interface. The user clicks this button or triggers the broadcast command by long-pressing the headphone touchpad. The APP calls the system's TTS engine (preferably using the offline engine, which is still available when there is no network) to generate PCM audio data. Subsequently, the APP sends the audio stream to the headphones via Bluetooth A2DP connection, and the headphones decode it and play it through the speaker. If the returned result is multilingual translated text, TTS automatically selects the corresponding language speaker. During the broadcast, playback automatically pauses when the user removes the headphones and resumes when the user puts them back on.

[0069] Example 3: Taking a user recording while only wearing the wireless earbuds, with the charging case in standby or stored mode, and the mobile device (phone) near the user (e.g., in a pocket or on a table) as an example, this is the earbud-only recording mode. (Refer to...) Figure 4 As shown in the embodiments of this application, the real-time recording and transcription method in headphone-only recording mode includes: S101. Start the headphone-only recording mode and collect the first audio data.

[0070] The user triggers headphone-only recording in any of the following ways: Double-tap the touch-sensitive area of ​​the wireless earbuds (the capacitive touch chip detects two quick presses); Remotely start via the "Headphone Recording Only" button on the mobile app.

[0071] Upon startup, the microcontroller (MCU, such as Hengxuan BES2600) inside the wireless headset wakes up the audio codec and the first microphone (such as the FF1AMEMS microphone, with a signal-to-noise ratio ≥64dB). The headset begins to collect the user's near-field speech signal (such as the user's own speech, close-range conversation, or lecture content). The audio signal is converted into first audio data by the codec at a sampling rate of 16kHz, 16-bit quantization, and OPUS encoding (bit rate 32kbps). To reduce power consumption, only one microphone (the call microphone) remains operational inside the headset, and the active noise cancellation function is automatically turned off. The data is first temporarily stored in the headset's circular buffer (capacity 256KB, buffering approximately 64 seconds of audio).

[0072] S102, The first audio data is transmitted to the mobile terminal wirelessly.

[0073] The headset MCU reads the audio data from the buffer, packages it into a frame (approximately 800 bytes) every 200ms, and streams it to the mobile terminal in real time via classic Bluetooth (A2DP protocol, SBC encoding). To ensure transmission reliability, a separate RFCOMM channel is established between the headset and the mobile phone, and a sequence number and checksum are added to each data packet. If packet loss occurs due to interference, the headset will include a retransmission request in the next packet. This step corresponds to "the wireless headset sends the collected audio data to the mobile terminal via wireless transmission" in claim 1.

[0074] S103, The mobile terminal receives audio data and uploads it to the cloud.

[0075] The "Smart Recording" app on the mobile terminal receives audio frames via Bluetooth, decodes them, and reassembles them into a continuous PCM stream (16kHz mono). Every 2 seconds (or every 64KB of data), the app uploads the audio block to the cloud's large-scale model's speech recognition interface via HTTPS. Simultaneously, the app caches the original audio file locally (for retransmission in case of network interruption). This step corresponds to "the mobile terminal uploads audio data to the cloud for processing" in claim 1.

[0076] S104. The cloud processes the data and returns the results.

[0077] The cloud-based system deploys the Whisper-large-v3 ASR engine and the GPT-4o mini large language model. The cloud first performs speech recognition on the received audio blocks and returns timestamped text in real time (streaming output). Based on user presets in the app, the cloud can choose to perform the following additional processing: Text summary: When the user selects "Meeting Summary", the cloud receives each complete paragraph (separated by silent paragraphs or periods) and calls the large language model to generate key points, such as "User requirement: The project must be tested by Monday".

[0078] Multilingual translation: If the user sets the source language to Chinese and the target language to English, the cloud will translate the Chinese text immediately after outputting the Chinese transcription and return the English text.

[0079] The processing results are pushed to the mobile terminal APP in real time via WebSocket.

[0080] S105: The mobile terminal displays the processing results and can send voice feedback to the headset.

[0081] After receiving the results from the cloud, the app displays them in different areas on the screen: Transcription area: Displays the currently recognized text in a scrolling subtitle format.

[0082] Summary / Translation Area: Located below the transcription area, this area displays the generated summary or translation.

[0083] If the user has enabled the "Voice Feedback" function (the corresponding switch is turned on in the app), each time a complete transcribed text sentence is received (separated by a period, question mark, or silence exceeding 0.5 seconds), the app calls a local TTS engine (such as Android's TextToSpeech, using an offline voice pack) to synthesize the text into natural speech and sends it to the wireless headset in real time via Bluetooth SCO link. The headset speaker plays the speech, allowing the user to hear the transcribed content without looking at the screen. During playback, if the headset microphone detects the user speaking (Voice Activity Detection (VAD), TTS automatically pauses and resumes half a second after the user stops speaking, avoiding interruption to the user's speech.

[0084] S106. Recording ends and data is saved.

[0085] The user double-tap the earphone touch area again or tap "Stop" via the app, and the earphones will stop recording and send an end signal. The mobile device will package and save the complete recording file (raw audio) and all processing results (JSON format) returned from the cloud in a local database, which the user can play back, edit, or share at any time.

[0086] Example 4: This example illustrates a scenario where the user records audio using only the charging case, with the wireless earbuds either stored away or not worn (e.g., the charging case is placed on a conference table and operates independently), and the mobile device (phone) is located in the user's pocket, backpack, or office desk. This is known as charging case-only recording mode. (Reference) Figure 5 As shown in the embodiments of this application, the real-time recording and transcription method in charging case-only recording mode includes: S201. Start the charging case-only recording mode and collect the second audio data.

[0087] The user can trigger charging case-only recording in any of the following ways: Press and hold the physical button on the side of the charging case (for example, for 2 seconds), and the built-in RGB indicator light in the charging case will flash red to indicate that recording has started.

[0088] Recording can be remotely started via the "Charging Case Recording Only" button on the mobile app (the charging case must be paired with the phone via Bluetooth).

[0089] Upon startup, the main control chip (MCU, such as Realtek RTL8763E) inside the charging case wakes up the second microphone (dual MEMS array, supporting beamforming) and audio processing unit. The charging case begins to collect ambient far-field speech (such as multiple speakers in a conference room, classroom lectures, and interview conversations). The audio signal is converted into second audio data by a codec at a sampling rate of 16kHz or 48kHz (user-configurable), 16-bit quantization, and OPUS encoding (bit rate adjustable from 32-192kbps). The data is written in real time to the charging case's internal high-capacity storage unit (e.g., 16GB eMMC). Simultaneously, the charging case maintains a circular buffer (512KB) in memory to temporarily store the most recent 10 seconds of audio for real-time uploading.

[0090] S202, The charging case transmits the second audio data to the mobile terminal wirelessly (adaptively selects WiFi or Bluetooth).

[0091] The charging case integrates a WiFi module (2.4GHz 802.11n) and a Bluetooth 5.2 module. During or after recording, the charging case needs to send audio data to the mobile terminal. The system automatically selects the optimal transmission method based on the characteristics of the audio data and current network conditions. The specific judgment logic is as follows: Judgment criteria: If the user selects "High Quality Mode" (sampling rate ≥ 44.1kHz or bit rate ≥ 192kbps), or recording duration ≥ 10 minutes, or file size ≥ 50MB, the data is judged as "High Quality / Long Duration". Otherwise (sampling rate ≤ 16kHz and bit rate ≤ 64kbps and recording duration < 10 minutes and file size < 50MB), the data is judged as "Low Quality / Short Duration".

[0092] Transmission Strategy: High-Quality / Long-Duration Data: The charging case prioritizes WiFi transmission. It first scans for known WiFi networks (such as home, office, or personal hotspots enabled by the phone). If an available WiFi network is found, the audio file is uploaded in chunks (1MB each) via HTTP POST to the receiving address specified on the mobile device. WiFi transmission speeds can reach 10-50Mbps, enabling fast transfer of large files. If no available WiFi network is found, the charging case sends a prompt to the phone via Bluetooth suggesting a WiFi connection for transferring large files, and waits for user input; the user can choose to ignore the prompt and continue with Bluetooth transmission (but at a slower speed). Low-Quality / Short-Duration Data: The charging case directly uses Classic Bluetooth (OBEX Object Push Protocol) or Bluetooth Low Energy (ATT File Transfer) to send audio files to the mobile device. Bluetooth transmission speeds are approximately 50-100KB / s, allowing small files of a few MB to be transferred within tens of seconds.

[0093] Transmission Timing: Users can set "Real-time Transmission" (transmit while recording) or "Transmit after recording is complete" in the app. In real-time transmission mode, the charging case sends a data packet every 2 seconds of accumulated audio (approximately 8-48KB, depending on the bitrate); in complete transmission mode, the charging case only starts transmitting after recording stops.

[0094] S203, The mobile terminal receives the second audio data and uploads it to the cloud.

[0095] The app on the mobile device receives a second audio data packet from the charging case via WiFi or Bluetooth, reassembles it into a complete audio stream (if it's real-time transmission, it's spliced ​​sequentially; if it's file transfer, it's stored directly). The app then uploads the audio data in chunks to the cloud's large-scale audio interface via HTTPS / WebSocket protocol. The default upload chunk size is 2 seconds of audio (approximately 8-48KB), automatically adjusting to 5-second chunks when the network is poor.

[0096] S204. The cloud processes the data and returns the results.

[0097] The cloud-based system deploys the Whisper-large-v3 ASR engine and the GPT-4o mini large language model. The cloud first performs speech recognition on the audio and returns timestamped text in real time. Based on user presets in the app (claim 2), the cloud can choose to perform text summarization and / or multilingual translation. The processing results are pushed to the mobile app in real time via WebSocket.

[0098] S205. The mobile terminal displays the processing results.

[0099] After receiving the results from the cloud, the app displays them in different areas on the screen: Transcription area: Displays the currently recognized text sentence by sentence in a scrolling caption format. Summary / Translation area: Displays the generated summary or translation. Users can click on any section of text to copy, share, or save it as a note.

[0100] S206. The mobile terminal sends the result back to the wireless headset.

[0101] If the user is wearing wireless headphones (e.g., simultaneously listening to the transcription results), the mobile terminal can synthesize the processing results into speech and send it to the wireless headphones for playback via Bluetooth. The specific implementation method is similar to step S105 in Embodiment 3, and will not be repeated here.

[0102] S207: Recording ends and data is saved.

[0103] The user presses and holds the charging case button again or taps "Stop" via the app to stop recording and send an end signal. The charging case saves the complete recording file (original audio) and transmission log locally. The mobile device associates the processing results returned from the cloud with the original audio and stores them, allowing the user to play back, edit, or share at any time.

[0104] Example 5: The system simulates a scenario where a user simultaneously wears wireless earbuds and places the charging case nearby (e.g., on a meeting room table). Both devices simultaneously capture audio from different sound fields: the earbuds capture near-field speech (the user's own speech or nearby sounds), while the charging case captures far-field environmental speech (other people in the meeting and background noise). The system supports seamless switching: when the user places the earbuds into the charging case, recording continues uninterrupted, and the charging case automatically takes over and receives the recorded data from the earbuds – this is the recording mode. (Reference) Figure 6 As shown in the embodiments of this application, the method for collaborative recording and seamless switching in simultaneous recording mode includes: S301. Start simultaneous recording mode (collaborative mode).

[0105] Users can trigger the recording function by clicking the "Collaborative Recording" button on the mobile app or by using a combination of physical buttons on the charging case (e.g., double-press followed by a long press). The app sends a start command via Bluetooth to both the earbuds and the charging case, including parameters such as synchronization timestamp, sampling rate (default 16kHz), and encoding format (OPUS, 32kbps). Upon receiving the command, both the earbuds and the charging case activate their microphones and audio codecs and begin collecting audio data. The earbuds collect the first audio data (near field), and the charging case collects the second audio data (far field). To ensure alignment between the two audio streams, the charging case acts as the master clock, broadcasting the current Bluetooth clock count via BLE. Upon receiving this, the earbuds synchronize their timers to the charging case clock, with a synchronization accuracy of ≤±1ms.

[0106] The S302 earphones and charging case record audio simultaneously and back it up in real time.

[0107] During recording, the earbuds send a compressed packet of their first audio data (approximately 1 second of audio, 4KB) to the charging case via Bluetooth Low Energy every second for real-time backup. Upon receiving the backup data, the charging case temporarily stores it in the "earbud backup area" within its internal dual-buffered system. Simultaneously, the charging case itself writes its second audio data to the "charging case recording area." The two areas employ a ping-pong buffering mechanism to prevent write conflicts. This backup process ensures that even if the earbuds are unexpectedly powered off or damaged, the recorded near-field audio will not be lost.

[0108] S303, Test by placing the earphones into the charging case.

[0109] When the user places the wireless earbuds into the charging slot of the charging case, the placement detection unit (Hall sensor or reed switch) inside the charging case immediately generates an interrupt signal. After the charging case MCU captures this signal, it continuously samples for 50ms to confirm that the state is stable, and checks the charging contact current (>10mA) to further confirm that the earbuds are placed correctly.

[0110] S304: The charging case sends a stop command, and the earbuds stop recording and transmit the recorded data.

[0111] Upon confirmation of placement in the charging case, the charging case sends a "Stop Recording" command (command code 0x01) to the earbuds via Bluetooth Low Energy. Upon receiving the command, the earbuds stop microphone sampling and turn off the audio codec power within 10ms to conserve power. Simultaneously, the earbuds transmit all untransmitted initial audio data (including the last few seconds of real-time backup) from their internal buffer to the charging case via BLE file transfer. An automatic retransmission mechanism ensures data integrity.

[0112] S305, the charging case continues to record continuously.

[0113] During the process of sending the stop command and receiving data from the earphones, the charging case's second microphone remains active, continuously capturing ambient voice. The charging case's MCU employs dual-thread processing: the main thread handles recording and writing to Flash, while the secondary thread handles communication with the earphones and data reception. Since the earphone data transmission rate (approximately 50KB / s) is much lower than the recording data writing rate, the two can operate in parallel without conflict. Recording is completely seamless, without any dropped frames or interruptions.

[0114] S306, Data Packaging and Merging.

[0115] After the user finishes recording (e.g., by pressing the charging case button again), the charging case performs a data packaging process. The packaging method is preset by the user in the app: Merged files: The first audio data (from the headphones) and the second audio data (from the charging case) are encoded into a stereo WAV file, with the left channel representing the near field and the right channel representing the far field. The sampling rate is uniformly resampled to 16kHz, with 16-bit quantization. The original timestamp and sync offset are appended to the file header.

[0116] Independent files: Save them as two separate MP3 files, with filenames such as "20250616_143000_ear.mp3" and "20250616_143000_env.mp3", and generate a JSON metadata file (recording the recording start time, duration, headphone serial number, etc.).

[0117] After packaging, the charging case adaptively selects the transmission method based on data characteristics and network conditions: if the total file size is ≥50MB or the user marks it as "high quality", WiFi transmission is used first; otherwise, Bluetooth transmission is used.

[0118] S307, the charging case sends the packaged file to the mobile terminal.

[0119] The charging case sends the packaged files (merged files or individual files + metadata) to the mobile device via WiFi or Bluetooth. If WiFi is used, the charging case automatically connects to a known WiFi network or mobile hotspot and uploads in chunks (1MB each, supporting resumeable upload) via HTTP POST. If Bluetooth is used, the OBEX object push protocol is employed. Upon completion of the transfer, the mobile app sends a confirmation receipt to the charging case, which can then delete local files to free up space.

[0120] S308: The mobile terminal uploads the data to the cloud and displays the processing results.

[0121] After receiving the audio file sent by the charging case, the mobile app determines the upload strategy based on the file type (merged or independent): For merged files: upload directly to the cloud, where the left and right channels can be processed separately.

[0122] For individual files: they can be uploaded separately, or they can be merged into a 3D stream and then uploaded.

[0123] The cloud performs speech recognition to text conversion, text summarization, and multilingual translation, with the results returned to the app in real time via WebSocket. The app displays the transcribed content and summary / translation on the interface. Notably, the app can present the processing results of the two audio streams differently: for example, displaying "Near Field Key Points" (a summary of the headphone user's individual remarks) on the left and "Far Field Meeting Minutes" (the overall discussion content) on the right.

[0124] S309, The mobile terminal sends the result back to the wireless headset.

[0125] If the user wears the wireless earphones again later, the mobile terminal can synthesize the processing results (such as summary text) into speech and send it to the earphones via Bluetooth for playback, making it convenient for the user to listen to the meeting minutes without checking their phone.

[0126] S310, End of recording and data saving.

[0127] After the user stops recording via the app or the charging case button, the system packages and saves all original audio files, cloud processing results (JSON format), and metadata in the mobile device's local database. The user can play back, edit, share, or export the audio at any time.

[0128] In summary, compared with the prior art, the present invention has the following beneficial effects: Multi-directional audio acquisition enhances recording integrity: Audio can be acquired independently or simultaneously through both the wireless earbuds and the charging case, enabling dual-channel recording of near-field and far-field environmental speech. This meets the complete sound field requirements for various scenarios such as meetings, speeches, and interviews, avoiding information loss.

[0129] Collaborative recording and seamless switching ensure recording continuity: When the earbuds are placed in the charging case, the charging case automatically takes over the recording and receives the recorded data from the earbuds, enabling the recording task to continue continuously. This completely solves the problem of recording interruption caused by the earbuds being placed in the charging case, and avoids the loss of key audio segments.

[0130] Efficient wireless transmission, freeing you from wired constraints: Supports dual-mode wireless transmission via Bluetooth and WiFi, allowing users to export recording data anytime, anywhere without carrying a data cable; and adaptively selects the transmission method based on audio quality and duration (WiFi for high-quality / long-duration audio, Bluetooth for low-quality / short-duration audio), balancing transmission efficiency and power consumption.

[0131] Audio relay storage extends recording time and enhances data security: When the mobile terminal is unreachable, the earphones can back up the recording data to the charging case in real time via Bluetooth Low Energy. The charging case acts as a high-capacity storage relay, which not only solves the problems of small storage space and limited power consumption of the earphones, but also forms a double data protection to avoid recording loss.

[0132] Cloud-based intelligent processing enables instant conversion of recordings into information: It constructs an integrated closed loop of "recording - wireless transmission - cloud-based large-scale model processing - terminal display / voice feedback", automatically completing operations such as speech-to-text, text summarization, and multilingual translation. Users do not need to manually call third-party software, significantly improving work efficiency, especially suitable for scenarios such as meeting minutes and language learning.

[0133] Voice feedback enhances the interactive experience: Mobile terminals can synthesize text results processed in the cloud into speech and transmit them back to wireless headphones for playback, allowing users to obtain processed information in real time when it is inconvenient to look at the screen, thus improving the product's usability.

[0134] It is understood that the above scenarios are merely examples and do not constitute a limitation on the application scenarios of the technical solutions provided in the embodiments of this application. The technical solutions of this application can also be applied to other scenarios. For example, as those skilled in the art will know, with the evolution of system architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0135] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0136] The steps in the method of this application embodiment can be adjusted, combined, or deleted according to actual needs.

[0137] The units in the device of this application embodiment can be merged, divided, and deleted according to actual needs.

[0138] In this application, the same or similar terms, concepts, technical solutions and / or application scenario descriptions are generally described in detail only when they appear for the first time. When they appear again, they are generally not repeated for the sake of brevity. When understanding the technical solutions and other contents of this application, the same or similar terms, concepts, technical solutions and / or application scenario descriptions that are not described in detail later can be referred to their previous relevant detailed descriptions.

[0139] In this application, the descriptions of the various embodiments have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0140] The technical features of the present application can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the present application.

[0141] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for recording and transcribing audio, characterized in that, Applied to a system including wireless earphones, a charging case, and a mobile terminal, the method includes: The wireless earbuds and / or the charging case activate the recording mode to collect first audio data and / or second audio data respectively. The wireless earphones and / or the charging case will transmit the collected audio data to the mobile terminal wirelessly. The mobile terminal uploads the audio data to the cloud for processing, receives the processing results returned by the cloud, and displays them on the terminal interface.

2. The method according to claim 1, characterized in that, The cloud-based processing includes at least one of speech recognition to text, text summarization, and / or multilingual translation.

3. The method according to claim 1, characterized in that, The method further includes: During the simultaneous audio acquisition process of the wireless earbuds and the charging case, when it is detected that the wireless earbuds have been placed in the charging case; The charging case sends a stop recording command to the wireless earbuds, and the wireless earbuds stop recording and transmit the first audio data they have collected to the charging case for storage. The charging compartment continues to collect second audio data.

4. The method according to claim 3, characterized in that, The charging case detects whether the wireless earphones are placed inside the case via a reed switch or a Hall sensor.

5. The method according to claim 3, characterized in that, The charging case transmits the collected audio data to the mobile terminal wirelessly, including: The charging case packages the first audio data and the second audio data into a merged file or a separate file; The charging compartment transmits the merged file or the individual file to the mobile terminal wirelessly.

6. The method according to claim 1, characterized in that, The step of the wireless earphone transmitting the collected audio data to the mobile terminal via wireless transmission also includes: When the mobile terminal is out of wireless range or the wireless connection is unstable, the wireless earphones will send the first audio data to the charging case for caching. When the charging case detects that the wireless connection with the mobile terminal has been restored, it sends the first audio data to the mobile terminal.

7. The method according to claim 1 or 6, characterized in that, The wireless transmission method includes Bluetooth transmission and / or WiFi transmission; the step of the charging case sending the collected audio data to the mobile terminal also includes; When the second audio data collected by the charging case or the first audio data cached is high-quality audio data and / or long-duration audio data, WiFi transmission is used for transmission. When the second audio data collected by the charging case or the first audio data cached is low-quality audio data and / or short-duration audio data, Bluetooth transmission is used for transmission.

8. The method according to claim 1, characterized in that, The method further includes: after receiving the processing result returned by the cloud, the mobile terminal converts the processing result into voice and transmits it back to the wireless earphone via Bluetooth for playback.

9. A recording transcription system, characterized in that, The system includes: Wireless earphones, used to collect near-field speech as the primary audio data; The charging case is used to charge the wireless earbuds and also to collect far-field or ambient voice as second audio data, and to cache the first audio data. The mobile terminal is used to receive audio data sent by the wireless earphones and / or the charging case, send it to the cloud for processing, and display the processing results fed back from the cloud. The cloud-based system is used for speech recognition, text conversion, and intelligent processing of audio data, and the processing results are returned to the mobile terminal.

10. The system according to claim 9, characterized in that, The mobile terminal is also used to convert the processing result into voice information and send it to the wireless earphone for playback.

Images