Four-in-One: A Joint Approach to Inverse Text Normalization, Punctuation, Capitalization, and Disfluency for Automatic Speech Recognition.
Sharman TanPiyush BehreNick KibreIssac AlphonsoShuangyu ChangPublished in: CoRR (2022)
Keyphrases
- automatic speech recognition
- spontaneous speech
- conversational speech
- speech recognition
- spoken words
- speech signal
- hidden markov models
- broadcast news
- speech retrieval
- english text
- human machine interaction
- word error rate
- spoken language
- word recognition
- recognition errors
- acoustic features
- neural network
- length normalization
- noisy environments
- text mining
- information retrieval
- spoken document retrieval
- text retrieval
- speech corpus
- compound words
- image processing