Parallel feature extraction system and method for general specific voice in voice signal

Abstract

The invention discloses a parallel feature extraction system and method for general specific voice in a voice signal. A model mainly comprises a voice signal, an emotion recognition model, a voiceprint recognition model and a voice recognition model. Output features obtained by an LLD channel are combined with output features obtained by a spectrogram and a TEO channel to obtain 1*1024-dimensionalemotion features; the spectrogram is fed into a convolutional neural network CNN as an input, the spectrogram and MFCC are fed into Seq2Seq, the models of the two channels are fused and an attentionmechanism is add to form a voiceprint recognition model; 42-dimensional MFCC is adopted as input, and output is carried out in combination with a BIMLSTM channel and a Seq2Seq channel; the spectrogramis combined with a Seq2Seq channel for outputting; and the models of the two channels are fused and an attention mechanism is added to form a voice recognition model. The method has the advantages ofhigh accuracy, high integration level, capability of freely selecting schemes and the like.

Description

technical field [0001] The invention relates to the field of signal processing and extraction, in particular to a feature extraction system for speech tasks. Background technique [0002] Speech is the most effective, natural and important form of communication for human beings. To achieve communication between humans and machines through voice requires machines to be intelligent enough to recognize human voices. With the development of machine learning, neural network and deep learning theory, the degree of completion of speech recognition-related tasks is gradually improving, which is of great help to the computer to understand the content of speech. At present, speech recognition tasks mainly involve the following three recognition tasks: [0003] 1. Voiceprint recognition [0004] Voiceprint recognition, also known as speaker recognition, is a form of biometric recognition. It analyzes and processes the continuous speech signal of the speaker to extract discrete speech...

AI Technical Summary

Problems solved by technology

[0009] However, there are many defects or deficiencies in the actual application or design of the above three recognition tasks

For example: Voiceprint recognition, speech recognition, and emotion recognition task models are not common, input forms are not uniform, there is no universal solution, the integration accuracy is not high, the recognition accuracy rate of a single emotion recognition task is not high, etc.

Images