Non-linear representations, sensor reliability estimation and context-dependent fusion in the audiovisual recognition of speech in noise.
Pascal TeissierJean-Luc SchwartzAnne Guérin-DuguéPublished in: EUROSPEECH (1997)
Keyphrases
- context dependent
- noisy environments
- recognition engine
- multi sensor
- noisy speech
- semantic level
- sensor noise
- audio visual
- recognition rate
- speech recognition
- speech enhancement
- context free
- sensor fusion
- human face recognition
- speech signal
- automatic speech recognition systems
- object recognition
- data fusion
- low level
- noise reduction
- signal to noise ratio
- background noise
- emotion recognition
- sensor data
- sensor networks
- pattern recognition
- speech corpus
- natural language
- anti noise
- high level
- user centric
- extended gaussian image
- spoken words
- handwriting recognition
- recognition algorithm
- information fusion
- multimedia content
- activity recognition
- feature extraction
- text to speech
- automatic target recognition
- speaker identification
- broadcast news
- fusion method