Quantization Aware Training with Absolute-Cosine Regularization for Automatic Speech Recognition.
Hieu Duy NguyenAnastasios AlexandridisAthanasios MouchtarisPublished in: INTERSPEECH (2020)
Keyphrases
- automatic speech recognition
- speech recognition
- discriminative training
- speech signal
- broadcast news
- speech retrieval
- hidden markov models
- conversational speech
- word error rate
- acoustic models
- spoken words
- recognition errors
- word recognition
- spontaneous speech
- speech sounds
- speech corpus
- feature vectors
- speech synthesis
- noisy environments
- computer vision
- acoustic features
- image acquisition
- non stationary