Chinese speech proofreading method and device based on speech recognition

A speech recognition and speech recognition engine technology, applied in speech recognition, speech analysis, instruments, etc., can solve problems such as reducing work efficiency and increasing costs, and achieve the effect of improving proofreading efficiency and enhancing consistency
CN110534100AInactive Publication Date: 2019-12-03北京海天瑞声科技股份有限公司
10 Cites 7 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
北京海天瑞声科技股份有限公司
Publication Date
2019-12-03
Estimated Expiration
Not applicable · inactive patent

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention relates to the technical field of speech recognition, and provides a Chinese speech proofreading method and device based on speech recognition. The Chinese speech proofreading method based on speech recognition comprises the following steps: a text corpus obtaining step: obtaining a text corpus of an original text and corresponding text pinyin with tones; an audio acquisition step ofacquiring audio based on the text corpus; a recognition step: recognizing the audio through a voice recognition engine to obtain recognized characters and corresponding pinyin with tones; a comparison step: comparing the identification corpus information with the text information; and a proofreading step: according to a comparison result, identifying difference characters which are not matched with the text information in the identification corpus information and corresponding time periods of the difference characters in the audio. Through the comparison, the recognized text and the text corpus are proofread, so that the position of the difference text is identified, the position where a transliteration error occurs in the audio can be rapidly identified, the proofread efficiency is improved, and the consistency of the text and the audio is enhanced.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The present invention generally relates to the technical field of speech recognition, in particular to a Chinese speech proofreading method and device based on speech recognition. Background technique

[0002] In the production process of the Chinese speech synthesis library, after professional speakers are invited to complete the recording according to the designed text corpus, there will be problems such as omission, mispronunciation, over-reading and tone of voice, which will eventually lead to inconsistencies between the audio and the original text. The speech synthesis database has high requirements on the consistency of audio and text, and the expected accuracy rate can reach 99.9%. Therefore, after the recording is completed, proofreading has become an essential step. In the proofreading process, it is necessary to ensure a one-to-one correspondence between the audio and the marked text at the word level and phoneme level. Whether the audio and...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More