On the Relevance of Phoneme Duration Variability of Synthesized Training Data for Automatic Speech Recognition.
Nick RossenbachBenedikt HilmesRalf SchlüterPublished in: CoRR (2023)
Keyphrases
- automatic speech recognition
- training data
- speech recognition
- acoustic models
- hidden markov models
- word error rate
- speech signal
- broadcast news
- spoken words
- conversational speech
- recognition errors
- word recognition
- acoustic features
- noisy environments
- relevance feedback
- spontaneous speech
- speech corpus
- language model
- speaker adaptation
- phoneme recognition
- machine learning
- test collection
- speech recognizer
- non stationary
- computer vision
- information retrieval