Signal-to-Signal Ratio Independent Speaker Identification for Co-channel Speech Signals.
Rahim SaeidiPejman MowlaeeTomi KinnunenZheng-Hua TanMads Græsbøll ChristensenSøren Holdt JensenPasi FräntiPublished in: ICPR (2010)
Keyphrases
- speech signal
- speaker identification
- noisy environments
- non stationary
- speech recognition
- speech quality
- signal processing
- fundamental frequency
- speaker recognition
- sound signals
- automatic speech recognition
- audio signal
- frequency domain
- gaussian mixture model
- noisy images
- noise reduction
- generative model
- vocal tract
- speech enhancement
- spectral analysis
- em algorithm
- image compression
- language model
- machine learning