On the integration of time-frequency masking speech separation and recognition in underdetermined environments.
Ingrid JafariSerajul HaqueRoberto TogneriSven NordholmPublished in: ACSCC (2012)
Keyphrases
- recognition engine
- recognition accuracy
- recognition rate
- speech recognition
- speech corpus
- signal processing
- recognition algorithm
- object recognition
- pattern recognition
- automatic speech recognition systems
- speech signal
- real world
- continuous speech recognition
- spoken words
- digit recognition
- noisy environments
- character recognition
- wavelet transform
- audio visual
- speech recognition systems
- data integration
- frequency domain
- blind separation
- text recognition
- speaker recognition
- dialogue system
- recognition process
- activity recognition
- automatic recognition
- human activities
- human visual system
- hidden markov models
- multiresolution
- feature extraction