Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
AUDIO CONTENT RECOGNITION METHOD AND APPARATUS, AND DEVICE AND COMPUTER-READABLE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2022/037419
Kind Code:
A1
Abstract:
An audio content recognition method and apparatus (400), and an electronic device (500) and a computer-readable medium. The method comprises: segmenting audio (102) to obtain a speech segment set (103) and a non-speech segment set (104) (201, 301); determining the type and language information (105, 106) of each speech segment in the speech segment set (103) (202, 302); and for each speech segment in the speech segment set (103), performing speech recognition on the speech segment on the basis of the type and the language information (105, 106) of the speech segment, so as to obtain a first recognition result (203, 303). By means of recognizing speech and music segments in the audio (102) by using different models, better recognition effects can be achieved for both kinds of content of the audio (102). Furthermore, by means of using different models to recognize the audio (102) of different language content, the effect of speech recognition is further improved.

Inventors:
KONG YALU (CN)
HE YI (CN)
Application Number:
PCT/CN2021/110849
Publication Date:
February 24, 2022
Filing Date:
August 05, 2021
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
BEIJING BYTEDANCE NETWORK TECH CO LTD (CN)
International Classes:
G10L15/04; G10L15/00; G10L15/26; H04N21/233; H04N21/2343; H04N21/439; H04N21/4402
Foreign References:
CN111986655A2020-11-24
CN105845129A2016-08-10
CN106878805A2017-06-20
CN102881309A2013-01-16
CN110728976A2020-01-24
JP2010091675A2010-04-22
Attorney, Agent or Firm:
EAST & CONCORD PARTNERS (CN)
Download PDF: