Making Machines Understand Us in Reverberant Rooms: Robustness Against Reverberation for Automatic Speech Recognition.
Takuya YoshiokaArmin SehrMarc DelcroixKeisuke KinoshitaRoland MaasTomohiro NakataniWalter KellermannPublished in: IEEE Signal Process. Mag. (2012)
Keyphrases
- automatic speech recognition
- speech signal
- speech recognition
- hidden markov models
- sound source
- speech retrieval
- word error rate
- broadcast news
- noisy environments
- conversational speech
- non stationary
- spoken words
- vocal tract
- word recognition
- acoustic features
- recognition errors
- video sequences
- speech corpus
- spontaneous speech
- language model
- probabilistic model
- machine learning