Hooking up spectro-temporal filters with auditory-inspired representations for robust automatic speech recognition.
Bernd T. MeyerConstantin SpilleBirger KollmeierNelson MorganPublished in: INTERSPEECH (2012)
Keyphrases
- automatic speech recognition
- speech recognition
- noisy environments
- conversational speech
- speech retrieval
- speech signal
- broadcast news
- hidden markov models
- spoken words
- temporal information
- signal processing
- acoustic features
- word recognition
- spontaneous speech
- word error rate
- recognition errors
- speech corpus
- visual information