A Comparative Study on Selecting Acoustic Modeling Units in Deep Neural Networks Based Large Vocabulary Chinese Speech Recognition.
Xiangang LiYuning YangXihong WuPublished in: IScIDE (2013)
Keyphrases
- speech recognition
- speech recognition systems
- speaker independent
- neural network
- pattern recognition
- speech recognizers
- speech synthesis
- speech recognizer
- hidden markov models
- automatic speech recognition
- language model
- speech understanding
- acoustic models
- speech recognition technology
- noisy environments
- speech signal
- noisy speech
- speech processing
- speaker identification
- broadcast news
- acoustic features
- mel frequency cepstral coefficients
- visual features