Exploring Capabilities of Monolingual Audio Transformers using Large Datasets in Automatic Speech Recognition of Czech.
Jan LeheckaJan SvecAles PrazákJosef V. PsutkaPublished in: CoRR (2022)
Keyphrases
- automatic speech recognition
- cl sr
- speech retrieval
- broadcast news
- speech recognition
- cross language
- spontaneous speech
- spoken document retrieval
- conversational speech
- acoustic features
- document collections
- hidden markov models
- speech signal
- word error rate
- speaker identification
- question answering
- manually generated
- speech corpus
- spoken words
- information retrieval
- noisy environments
- multimedia
- cross language information retrieval
- text retrieval
- query expansion
- neural network
- word recognition
- audio visual
- text categorization
- language model
- relevance feedback
- image retrieval
- pattern recognition