Predicting speech recognition confidence using deep learning with word identity and score features.
Po-Sen HuangKshitiz KumarChaojun LiuYifan GongLi DengPublished in: ICASSP (2013)
Keyphrases
- speech recognition
- speech recognition systems
- deep learning
- unsupervised feature learning
- speech recognizer
- hidden markov models
- keyword spotting
- speech synthesis
- language model
- wall street journal corpus
- co occurrence
- speech signal
- automatic speech recognition
- speech recognizers
- low level
- noisy environments
- machine learning
- n gram
- unsupervised learning
- pattern recognition
- speaker identification
- feature vectors
- speaker independent
- speech retrieval
- feature extraction