Audio processing method and device

• Main Authors: SUN SINING, YANG ZHANHENG
• Format: Patent
• Language: Chinese; English
• Published: 27.10.2023
• Subjects: ACOUSTICS; MUSICAL INSTRUMENTS; PHYSICS; SPEECH ANALYSIS OR SYNTHESIS; SPEECH OR AUDIO CODING OR DECODING; SPEECH OR VOICE PROCESSING; SPEECH RECOGNITION

The invention provides an audio processing method and a related device. The embodiment of the invention can be applied to the field of artificial intelligence. The method comprises the steps that firstly, an audio signal is acquired, and the audio signal comprises N audio frames; secondly, the N audio frames are input into a streaming acoustic network, N phoneme features and N streaming audio features are obtained, and the N phoneme features are used for representing phoneme information of the audio signals; then, an entity set is obtained, the entity set comprises K entities which are constructed in advance, and the K entities correspond to the K pieces of phoneme information; then, L entities are extracted from the entity set according to the N phoneme features, and the L entities correspond to the N phoneme features; and finally, inputting the audio signal, the N streaming audio features and the L entities into a non-streaming acoustic network to obtain a text recognition result. According to the audio pro