Enhancing speech at very low signal-to-noise ratios using non-acoustic reference signals.
Ben MilnerPublished in: Speech Commun. (2013)
Keyphrases
- low signal to noise ratio
- acoustic signal
- acoustic signals
- random noise
- speech recognition systems
- speech sounds
- acoustic features
- speech recognition
- signal detection
- emotional speech
- low snr
- speech signal
- emotion recognition
- prosodic features
- acoustic models
- speaker identification
- text to speech
- audio visual
- spoken language
- fundamental frequency
- formant frequencies
- recognition engine
- bird species
- speaker independent
- underwater acoustic
- automatic speech recognition
- speech recognizers
- cepstral features
- vowel phonemes
- source localization
- spontaneous speech
- pattern recognition
- audio signals
- audio signal
- signal processing