Optimization of Speaker-Aware Multichannel Speech Extraction with ASR Criterion.
Katerina ZmolíkováMarc DelcroixKeisuke KinoshitaTakuya HiguchiTomohiro NakataniJan CernockýPublished in: ICASSP (2018)
Keyphrases
- automatic speech recognition
- speech recognition
- speech signal
- broadcast news
- hidden markov models
- speaker identification
- speaker verification
- noisy environments
- word error rate
- speaker recognition
- optimization problems
- audio visual
- optimization algorithm
- speech sounds
- spontaneous speech
- recognition errors
- speaker dependent
- speech corpus
- information extraction
- optimization criteria
- speech recognizer
- acoustic features
- speech synthesis
- vocal tract
- global optimization
- linear prediction
- conversational speech
- automatic speech recognition systems
- pattern recognition
- speaker diarization
- optimization process
- optimization method
- spoken words
- error rate
- prosodic features
- phoneme recognition
- speech segments
- multi modal
- information retrieval
- automatic transcription
- speech retrieval
- expectation maximization
- non stationary
- speaker independent
- mel frequency cepstral coefficients
- automatic extraction