Almost-unsupervised Speech Recognition with Close-to-zero Resource Based on Phonetic Structures Learned from Very Small Unpaired Speech and Text Data.
Yi-Chen ChenChia-Hao ShenSung-Feng HuangHung-yi LeeLin-Shan LeePublished in: CoRR (2018)
Keyphrases
- speech recognition
- text data
- text classification
- speech recognizer
- hidden markov models
- language model
- speech signal
- text mining
- speech processing
- speech synthesis
- structured data
- automatic speech recognition
- speaker identification
- speaker independent
- speech recognition technology
- noisy environments
- document collections
- speech recognition systems
- high dimensional
- pattern recognition
- high dimensional data
- unsupervised learning
- acoustic models
- isolated word
- text documents
- speech recognizers
- speaker diarization
- vocal tract
- information retrieval systems
- artificial intelligence
- speaker dependent
- information retrieval
- machine learning