Generative Adversarial Training Data Adaptation for Very Low-resource Automatic Speech Recognition.
Kohei MatsuuraMasato MimuraShinsuke SakaiTatsuya KawaharaPublished in: CoRR (2020)
Keyphrases
- automatic speech recognition
- training data
- speech recognition
- speaker adaptation
- discriminative training
- acoustic models
- speech signal
- hidden markov models
- word error rate
- conversational speech
- discriminative methods
- spoken words
- broadcast news
- decision trees
- speech retrieval
- noisy environments
- generative model
- learning algorithm
- recognition errors
- word recognition
- spontaneous speech
- speech corpus
- machine learning
- training process
- pattern recognition
- acoustic features
- image denoising
- speech recognizer
- non stationary
- information retrieval systems
- image processing
- feature selection