Maximum Entropy-Based Reinforcement Learning Using a Confidence Measure in Speech Recognition for Telephone Speech.
Carlos MolinaNéstor Becerra YomaFernando HuenupánClaudio GarretónJorge WuthPublished in: IEEE Trans. Speech Audio Process. (2010)
Keyphrases
- speech recognition
- confidence measure
- speech signal
- speech synthesis
- automatic speech recognition
- hidden markov models
- speech recognizer
- speaker identification
- language model
- speech processing
- pattern recognition
- confidence values
- speech recognition systems
- speech recognition technology
- noisy environments
- isolated word
- recognition engine
- speaker independent
- speech recognizers
- word error rate
- highly accurate
- acoustic models
- cepstral coefficients
- speech recognition errors
- speech retrieval
- noisy speech
- machine learning
- speaker adaptation
- image processing