A single speaker is almost all you need for automatic speech recognition.
Edresson CasanovaChristopher ShulbyAlexander KorolevArnaldo Cândido JúniorAnderson da Silva SoaresSandra M. AluísioMoacir Antonelli PontiPublished in: CoRR (2022)
Keyphrases
- automatic speech recognition
- speech recognition
- speech signal
- word error rate
- hidden markov models
- broadcast news
- noisy environments
- speech retrieval
- conversational speech
- acoustic features
- speech corpus
- recognition errors
- spontaneous speech
- word recognition
- spoken words
- neural network
- acoustic models
- speaker adaptation
- handwriting recognition