A highly concurrent voip recording service system based on intelligent speech recognition

Abstract

The invention relates to a high-concurrency VOIP recording service system based on intelligent voice recognition, and the system comprises a recording service module which is used for directly carrying out the decoding and storage of audio of left and right sound channels at equipment and a chip layer, and obtaining a recording audio; a buffer register which is configured on a software layer, timesynchronization is performed, and when audio concurrency higher than the current capacity occurs, the capacity of the buffer register is increased as required; a sound recording file management module which is used for inputting the sound recording audios decoded and stored by the decoding storage module into a buffer queue; a voice recognition engine which encodes and decodes the recorded audio,and assembles the voice media data packet into correct voice media stream data through feature extraction; and a background service system to which the voice media stream data are transmitted throughthe MQ message queue middleware, and the scheme not only can process more than 200 sessions at the same time, but also can improve the recording quality at the same time.

Description

technical field [0001] The invention relates to the field of recording services, in particular to a highly concurrent VOIP recording service system based on intelligent speech recognition. Background technique [0002] With the rapid development of IT technology, the traditional PSTN telephone network can no longer meet the needs of communication, especially after the emergence of VOIP, VoIP (Voice over Internet Protocol) in short is to digitize the analog signal (Voice) in the form of data packets ( Data Packet) is transmitted in real time on the IP network (IP Network). Enterprises adopt VOIP technology to gradually replace call center services based on PSTN lines to meet the needs of convenient, unified and cheap communication. However, with the demand brought by mobile Internet technology and industrial upgrading, traditional recording services have been difficult to meet customers' urgent demands for high concurrency and fast identification, and machine learning to rep...

AI Technical Summary

Problems solved by technology

However, with the demand brought about by mobile Internet technology and industrial upgrading, traditional recording services have been difficult to meet customers' urgent demands for high concurrency and fast identification, using machine learning to replace manpower in call centers with lower cost operations, etc.

[0003] The disadvantage of the existing technology is that in the current conventional solutions, the AI ​​engine is usually responsible for the recognition and decoding of audio input, feature extraction and then enters the recognition stage, and the decoding process itself consumes system resources and costs more time. Large, when there are multiple channels of voice input at the same time, the system resource consumption will be very huge, resulting in poor recording quality and poor communication. ability

Images