A method for shielding speech breathing noise

Abstract

The invention discloses a telephone transmitter respiration noise shielding method. The method comprises the following steps: according to noise features, a noise feature model is built; after pretreatment such as steady state noise removal and speech endpoint detection is carried out on acquired sound input signals, feature analysis and extraction are carried out, and a noise signal template in need of recognition is built; the already-existing noise template features and the inputted noise signal features are compared, and according to a certain search and pattern matching strategy, the best noise template is found out from a noise feature base to be subjected to matching treatment with the inputted noise; and a recognition result is given through a determination algorithm, and the corresponding noise is filtered. Noise acquired by the telephone transmitter is filtered through a suitable algorithm, the noise is removed, and no noise is heard from the amplified sound.

Description

technical field [0001] The invention relates to the technical field of speech signal processing, in particular to a breathing noise shielding method for speaking. Background technique [0002] At this stage, pilots of trainer planes, fighter jets and other aircraft strongly report that the sound of breathing in the mask is relatively heavy. In a small confined space, the sound of human breathing and the sound of airflow in the air pipe are picked up by the microphone and then amplified by multi-stage noise. , in the case of high load and low air pressure, people's breathing will be aggravated, which will intensify the intensity of the noise. After tower personnel and pilots receive the breathing sound, on the one hand, it will affect the communication quality of the helmet mask, and on the other hand, it will affect the listening quality. A certain degree of oppressive feeling will be caused by the personnel's psychology, which will cause psychological pressure and communica...

AI Technical Summary

Problems solved by technology

The breathing noise shielding technology of sending voice must establish a noise feature model according to the noise characteristics; compare the existing noise template features with the input noise signal features, and perform matching processing with the noise library; As a result, speech recognition technology builds its own speech library, which can only recognize speech, and the speech breathing noise shielding technology needs to build its own powerful noise library. Therefore, the existing speech recognition technology cannot meet the needs of the current stage, and speech breathing must be performed Research on Noise Shielding Technology

Images