Optimization of Speech Enhancement Front-End with Speech Recognition-Level Criterion.
Takuya HiguchiTakuya YoshiokaTomohiro NakataniPublished in: INTERSPEECH (2016)
Keyphrases
- speech recognition
- speech signal
- noisy environments
- speech enhancement
- noisy speech
- hidden markov models
- spectral subtraction
- speech synthesis
- automatic speech recognition
- vocal tract
- language model
- pattern recognition
- background noise
- signal to noise ratio
- speaker identification
- linear prediction
- speech recognition systems
- noise reduction
- single channel
- acoustic features