A Method for Automatically Learning and Optimizing Acoustic Models

Abstract

The invention discloses a method for automatically learning and optimizing an acoustic model. The problems that in the prior art, the acoustic model optimization process is tedious, time-consuming, high in cost and high in admission threshold are solved. The method comprises the following steps: S1, selecting part of annotation data from a database into a test pool, and screening the remaining part into a training pool; S2, training the data in the training pool in batches, performing loop iteration training by using the acoustic model with the highest recognition rate, and completing the optimal acoustic model by using the existing data; and S3, testing the completed optimal acoustic model by using the test pool, inputting a test result into a database, and generating a test report. According to the scheme, the annotation data are expanded through voice enhancement and other means, the manual annotation cost is reduced, and a key basis is provided for improving the accuracy of a voicerecognition result. And loop iteration is completed through automatic training and testing, time consumption and cost of an acoustic model optimization process are reduced, and an admission thresholdis reduced.

Description

technical field [0001] The invention relates to the field of computers, in particular to a method for automatically learning and optimizing an acoustic model. Background technique [0002] Speech recognition technology is to enable smart devices to understand human speech. It is a multidisciplinary science involving digital signal processing, artificial intelligence, linguistics, mathematical statistics, acoustics, emotion and psychology. This technology can provide many applications such as automatic customer service, automatic voice translation, command control, voice verification code, etc. In recent years, with the rise of artificial intelligence, speech recognition technology has made great breakthroughs in both theory and application. It has begun to move from the laboratory to the market, and has gradually entered our daily life. Now speech recognition has been used in many fields, mainly including speech recognition dictation, voice paging and answering platform, i...

AI Technical Summary

Problems solved by technology

[0008] The present invention mainly solves the problems of cumbersome and time-consuming process of optimizing the acoustic model in the prior art, high cost and high barriers to entry; provides a method for automatically learning and optimizing the acoustic model, automatic training, automatic testing and iterative optimization, reducing acoustic model optimization Time-consuming and cost-intensive process, lowering barriers to entry

Images