18258316. SPEECH RECOGNITION METHOD AND APPARATUS simplified abstract (HUAWEI TECHNOLOGIES CO., LTD.)
Contents
SPEECH RECOGNITION METHOD AND APPARATUS
Organization Name
Inventor(s)
SPEECH RECOGNITION METHOD AND APPARATUS - A simplified explanation of the abstract
This abstract first appeared for US patent application 18258316 titled 'SPEECH RECOGNITION METHOD AND APPARATUS
Simplified Explanation
The patent application describes a method and apparatus for speech recognition. Here is a simplified explanation of the abstract:
- A terminal device inputs a phoneme to be recognized into a multitask neural network model.
- The neural network model outputs a prediction result, which includes both a character prediction and a punctuation prediction corresponding to the input phoneme.
- The terminal device displays at least a part of the prediction result on its display.
Potential applications of this technology:
- Speech recognition systems in various devices such as smartphones, tablets, and computers.
- Voice-controlled virtual assistants.
- Transcription services for converting spoken language into written text.
Problems solved by this technology:
- Simultaneous prediction of both characters and punctuation marks corresponding to a phoneme.
- Efficient and accurate speech recognition.
- Reducing the size of the neural network model for deployment on terminal devices.
Benefits of this technology:
- Improved user experience with accurate and real-time speech recognition.
- Enhanced productivity through voice-controlled applications.
- Reduced computational resources required for speech recognition on terminal devices.
Original Abstract Submitted
This application relates to a speech recognition method and apparatus. The speech recognition method includes: A terminal device inputs a to-be-recognized phoneme into a first multitask neural network model; the first multitask neural network model outputs a first prediction result, where the first prediction result includes a character prediction result and a punctuation prediction result that correspond to the to-be-recognized phoneme; and the terminal device displays at least a part of the first prediction result on a display of the terminal device. A neural network model for simultaneously predicting a character and a punctuation corresponding to a phoneme is constructed, so that the character and the punctuation corresponding to the phoneme can be simultaneously output. In addition, the neural network model is small-sized, and can be deployed on a terminal side.