Quartznet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutions.
Samuel KrimanStanislav BeliaevBoris GinsburgJocelyn HuangOleksii KuchaievVitaly LavrukhinRyan LearyJason LiYang ZhangPublished in: ICASSP (2020)
Keyphrases
- automatic speech recognition
- speech recognition
- speech signal
- hidden markov models
- noisy environments
- conversational speech
- word error rate
- spoken words
- word recognition
- multi channel
- broadcast news
- recognition errors
- multi modal
- image compression
- speech corpus
- speech retrieval
- discriminative training
- acoustic features
- spontaneous speech
- wavelet transform