A comparison of broad phonetic and acoustic units for noise robust segment-based phonetic recognition.
Tara N. SainathVictor ZuePublished in: INTERSPEECH (2008)
Keyphrases
- speech recognition
- speaker independent
- speech sounds
- noisy environments
- noisy speech
- speech recognition systems
- hidden markov models
- speaker identification
- emotional speech
- image noise
- acoustic models
- speech signal
- noise level
- speech recognizer
- speech segments
- robust recognition
- recognition rate
- salt pepper
- automatic speech recognition
- spoken term detection
- partial occlusion
- recognition accuracy
- object recognition
- pose invariant face recognition
- random noise
- vocal tract
- greater robustness
- geometric distortions
- estimation error
- word level
- spoken language
- information retrieval
- registration errors
- automatic recognition
- anti noise
- median filter
- imaging artifacts
- activity recognition
- language model
- pattern recognition
- image processing