Method of converting whispered voice into normal voice based on radial group neutral network

Abstract

The invention discloses a method of converting whispered voice into normal voice based on a radial group neutral network, which comprises two steps of training and converting: respectively extracting line spectrum pair parameters of the whispered voice and the normal voice when in training; seizing mapping relation of spectrum envelope of the whispered voice and the normal voice by using the radial group neutral network; preprocessing the whispered voice when in conversion and extracting the line spectrum pair parameters, converting the line spectrum pair parameters of the whispered voice by the trained radial group neutral network, at last generating a driving source of the voice by using basic frequency mean value of the voice as basic voice frequency, and converting into the normal voice by a line spectrum pair synthesizer. The whispered voice converted by the invention achieves better effect on the respects of intelligibility threshold and tone quality.

Description

technical field [0001] The invention belongs to the technical field of speech signal processing, in particular to the technology of converting ear speech into normal speech. Background technique [0002] Otospeech is a pronunciation pattern that differs from normal speech and is characterized by low volume and no vocal cord vibration at all. When speaking on certain occasions, people often use whispering in order not to affect others or to keep the conversation confidential. However, due to the particularity of whispering pronunciation and the influence of the dialogue environment, the voice signal in this way not only has a low signal-to-noise ratio but also has poor intelligibility and clarity, especially when communicating through communication equipment, which affects the conversation quality, and it is easy to cause fatigue. In addition, some voice patients or those with abnormal pronunciation can only use whispering when communicating, which affects communication. ...

AI Technical Summary

Problems solved by technology

However, due to the particularity of whispering pronunciation and the influence of the dialogue environment, the voice signal in this way not only has a low signal-to-noise ratio but also has poor intelligibility and clarity, especially when communicating through communication equipment, which affects the conversation Quality, but also easy to cause fatigue

In addition, some voice patients or those with abnormal pronunciation function can only use whispering when communicating, which affects communication.

[0003] At present, there are few studies on ear-to-speech conversion at home and abroad. The existing methods are: 1. Use the linear prediction method (LPC) to realize the reconstruction of the ear-to-speech, and synthesize the oblique lattice synthesis filter by extracting the partial correlation coefficient of the ear-to-speech , the problems encountered are: firstly, the formant of ear speech is not easy to extract; 2. Using the Mixed Excitation Linear Prediction Model (MELP) to reconstruct ear speech, the speech is divided into five frequency bands, and four low-frequency bands are used as voiced segments 3. Based on the homomorphic signal processing method combined with the relative entropy phonological segmentation whisper reconstruction system, after the earphone phonological segmentation, use the homomorphic signal The processing method obtains the vocal tract response sequence, and adds the fundamental frequency according to the tone. Since the ear speech transfer function is different from the normal speech, although some post-processing has been done, the naturalness of the converted speech is not high.

Images