Chinese dialect language classification system based on parameter sparse sharing

• Main Authors: REN JIE, TIAN FENG, WAN MIAO, MA MIAO
• Format: Patent
• Language: Chinese; English
• Published: 30.07.2021
• Subjects: ACOUSTICS; CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; MUSICAL INSTRUMENTS; PHYSICS; SPEECH ANALYSIS OR SYNTHESIS; SPEECH OR AUDIO CODING OR DECODING; SPEECH OR VOICE PROCESSING; SPEECH RECOGNITION

A Chinese dialect language classification system based on parameter sparse sharing comprises a feature extraction module, a basic network training module, a subtask network training module, a joint training module and an output module. A voice feature Mel frequency cepstrum coefficient is extracted, a basic network is constructed based on different voice features, subtasks are defined, and network parameters with optimal test set accuracy for the subtasks are selected based on the basic network; each sub-task network is integrated into a model based on a parameter sparse sharing mechanism, and corresponding sub-task network parameters in the basic network are updated by using each sub-task data set pair; the model based on the parameter sparse sharing mechanism has the advantages of high classification accuracy and simplified network parameters. The speech features are input into a trained model based on a parameter sharing mechanism, and a prediction result of a dialect language can be obtained. 基于参数稀疏共享的中国方