Semi-Supervised Training of Transformer and Causal Dilated Convolution Network with Applications to Speech Topic Classification
Open Access
- 20 June 2021
- journal article
- research article
- Published by MDPI AG in Applied Sciences
- Vol. 11 (12) , 5712
Abstract
Aiming at the audio event recognition problem of speech recognition, a decision fusion method based on the Transformer and Causal Dilated Convolutional Network (TCDCN) framework is proposed. This method can adjust the model sound events for a long time and capture the time correlation, and can effectively deal with the sparsity of audio data. At the same time, our dataset comes from audio clips cropped by YouTube. In order to reliably and stably identify audio topics, we extract different features and different loss function calculation methods to find the best model solution. The experimental results from different test models show that the TCDCN model proposed in this paper achieves better recognition results than the classification using neural networks and other fusion methods.Keywords
Funding Information
- National Office for Philosophy and Social Sciences (17BTQ062)
This publication has 21 references indexed in Scilit:
- Sound Event Detection of Weakly Labelled Data With CNN-Transformer and Automatic Threshold OptimizationIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2020
- Automatic meeting summarization and topic detection systemData Technologies and Applications, 2018
- Audio Set Classification with Attention Model: A Probabilistic Perspective2018
- Multiple topic identification in human/human conversationsComputer Speech & Language, 2015
- Topic identification techniques applied to dynamic language model adaptation for automatic speech recognitionExpert Systems with Applications, 2015
- Unsupervised training of an HMM-based self-organizing unit recognizer with applications to topic classification and keyword discoveryComputer Speech & Language, 2014
- Statistical language model adaptation: review and perspectivesSpeech Communication, 2004
- Speaker-independent phonetic classification using hidden Markov models with mixtures of trend functionsIEEE Transactions on Speech and Audio Processing, 1997
- Speaker-independent phone recognition using hidden Markov modelsIEEE Transactions on Acoustics, Speech, and Signal Processing, 1989
- An Introduction to the Application of the Theory of Probabilistic Functions of a Markov Process to Automatic Speech RecognitionBell System Technical Journal, 1983