Audio-to-text alignment for speech recognition with very limited resources.
Xavier AngueraJordi LuqueCiro GraciaPublished in: INTERSPEECH (2014)
Keyphrases
- limited resources
- speech recognition
- speaker identification
- speech processing
- speech recognition technology
- speech synthesis
- hidden markov models
- text to speech
- audio visual speech recognition
- speech recognizer
- language model
- automatic speech recognition
- cepstral coefficients
- information retrieval
- multimedia
- speech recognition systems
- word level
- speech signal
- noisy environments
- handwriting recognition
- pattern recognition
- signal processing
- visual information
- speaker dependent
- gaussian mixture model
- visual data
- audio visual
- broadcast news
- conversational speech
- students with learning disabilities
- video search
- speech recognizers
- bayesian networks