Television drama audio segmentation method, device and equipment

• Main Authors: CHEN JIANCHAO, XIAO LONGYUAN, CAI ZHENHUA, LIU XIAOWEI, LI XIMIN
• Format: Patent
• Language: Chinese; English
• Published: 12.06.2020
• Subjects: ACOUSTICS; MUSICAL INSTRUMENTS; PHYSICS; SPEECH ANALYSIS OR SYNTHESIS; SPEECH OR AUDIO CODING OR DECODING; SPEECH OR VOICE PROCESSING; SPEECH RECOGNITION

The invention discloses a television drama audio segmentation method, device and equipment. The method comprises the following steps of adopting a baseline system to carry out original segmentation ontelevision drama audios to obtain original segmentation results, adopting a semi-supervised automatic speech recognition mode, carrying out the automatic speech recognition on each segment of audio in the original segmentation result, to obtain speech recognition coding results, and employing a dynamic planning approach, aligning and comparing the speech recognition coding result with an originalannotation text associated with the television drama audio; and calculating a word matching rate between the original segmentation result and the acoustic model, re-segmenting each segment of audio in the original segmentation result according to the word matching rate, updating the acoustic model according to each segment of audio after re-segmentation, and aligning the voice segment of the television drama audio with th