Deep learning-based voice superposition detection method and device, and computer equipment

• Main Authors: FENG QINYU, ZHAO XIANGYU, HUANG WENQI, LIN QUANCHEN, YAO SENJING, LIANG LINGYU
• Format: Patent
• Language: Chinese; English
• Published: 28.03.2023
• Subjects: ACOUSTICS; CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; MUSICAL INSTRUMENTS; PHYSICS; SPEECH ANALYSIS OR SYNTHESIS; SPEECH OR AUDIO CODING OR DECODING; SPEECH OR VOICE PROCESSING; SPEECH RECOGNITION

The invention relates to a voice superposition detection method and device based on deep learning, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring a to-be-recognized voice signal, and performing voice feature extraction on the to-be-recognized voice signal to obtain a voice feature corresponding to each voice frame in the to-be-recognized voice signal; for any voice frame, inputting the voice feature corresponding to the voice frame and the voice feature corresponding to a target voice frame associated with the voice frame into a pre-trained neural network probability model to obtain a voice frame category probability vector corresponding to the voice frame; according to the voice frame category probability vector corresponding to each voice frame, determining a to-be-recognized sequence corresponding to the to-be-recognized voice signal; and according to the to-be-recognized sequence, determining a coincident tone segment in the to-be-r