Real-time speech stream extraction and speech recognition system and method

Abstract

The embodiment of the invention discloses a speech stream extraction and speech recognition system and a speech stream extraction and speech recognition method. The system comprises an audio collectorand an automatic speech recognition module. The audio collector is connected with the automatic speech recognition module. The audio collector is arranged on a first data line and a second data line.The first data line is connected with conversation equipment and a microphone. The second data line is connected with the conversation equipment and a loudspeaker. The audio collector is used for collecting a first audio sent by the microphone to the conversation equipment. The audio collector is also used for collecting a second audio sent by the conversation equipment to the loudspeaker. The audio collector sends the first audio and the second audio to the automatic speech recognition module. After the automatic speech recognition module carries out speech recognition on the first audio andthe second audio, a recognition result is sent to the user terminal to be displayed. The embodiment of the invention has the advantages that resources are saved. The system docking operation is simple and convenient, and the ASR identification engine can be used at any time.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of artificial intelligence and big data analysis, and in particular to a real-time voice stream extraction and voice recognition system and method. Background technique [0002] Automatic Speech Recognition (ASR, Automatic Speech Recognition) is a technology that converts human speech into text. Speech recognition is a multidisciplinary field, which is closely connected with many disciplines such as acoustics, phonetics, linguistics, digital signal processing theory, information theory, and computer science. Due to the diversity and complexity of speech signals, speech recognition systems can only obtain satisfactory performance under certain restrictions, or can only be applied to certain specific occasions. The performance of a speech recognition system roughly depends on the following four types of factors: 1. The size of the recognition vocabulary and the complexity of the speech; ...

AI Technical Summary

Problems solved by technology

[0004] To this end, the embodiment of the present invention provides a real-time voice stream extraction and voice recognition system and method to solve the cumbersome and time-consuming problems in the prior art because the ASR voice recognition engine needs to be configured on the corresponding server.

Images