SPEECH DATA RECOGNITION METHOD, DEVICE, AND MEDIUM

• Main Author: SONG, Yuanfeng
• Format: Patent
• Language: Chinese; English; French
• Published: 18.11.2021
• Subjects: ACOUSTICS; MUSICAL INSTRUMENTS; PHYSICS; SPEECH ANALYSIS OR SYNTHESIS; SPEECH OR AUDIO CODING OR DECODING; SPEECH OR VOICE PROCESSING; SPEECH RECOGNITION

A speech data recognition method and apparatus, a device, and a medium. The method comprises: performing speech recognition on speech data to be recognized, so as to obtain candidate results of said speech data (S10); acquiring an initial sorting result of the candidate results, and acquiring associated subject information of the candidate results (S20); re-sorting the candidate results on the basis of the initial sorting result and the associated subject information, so as to obtain a target sorting result (S30); and selecting a target candidate result from among the candidate results according to the target sorting result, and taking the selected target candidate result as a speech recognition result of said speech data (S40). La présente invention concerne un procédé et un appareil de reconnaissance de données vocales, un dispositif et un support. Le procédé consiste : à effectuer une reconnaissance vocale sur des données vocales à reconnaître de sorte à obtenir des résultats candidats desdites données voc