A path-based layered architecture using HMM for automatic speech recognition.
Marta CasarJosé A. R. FonollosaAlbino NogueirasPublished in: EUSIPCO (2006)
Keyphrases
- automatic speech recognition
- layered architecture
- hidden markov models
- speech recognition
- speech signal
- conversational speech
- agent architecture
- spoken words
- word error rate
- speaker adaptation
- broadcast news
- speech corpus
- noisy environments
- recognition errors
- discriminative training
- spontaneous speech
- web services
- speech retrieval
- acoustic features
- word recognition
- artificial intelligence
- vocal tract
- speech synthesis
- speaker independent
- neural network
- gesture recognition
- non stationary
- multiscale