A noise-type and level-dependent MPO-based speech enhancement architecture with variable frame analysis for noise-robust speech recognition.
Vikramjit MitraBengt J. BorgstromCarol Y. Espy-WilsonAbeer AlwanPublished in: INTERSPEECH (2009)
Keyphrases
- noisy environments
- speech recognition
- speech enhancement
- noisy speech
- spectral subtraction
- speech signal
- noise reduction
- background noise
- hidden markov models
- automatic speech recognition
- speaker identification
- pattern recognition
- spectral analysis
- language model
- noise model
- speech synthesis
- vocal tract
- signal to noise ratio
- speaker recognition
- speaker diarization
- smoothing algorithm
- speech recognition systems