Deriving Phonetic Transcriptions and Discovering Word Segmentations for Speech-to-Speech Translation in Low-Resource Settings.
Andrew WilkinsonTiancheng ZhaoAlan W. BlackPublished in: INTERSPEECH (2016)
Keyphrases
- spoken document retrieval
- speech recognition
- spoken term detection
- broadcast news
- speech recognizer
- spoken documents
- automatic speech recognition
- spontaneous speech
- out of vocabulary
- speech recognition systems
- cross language
- word error rate
- speech sounds
- automatic transcription
- spoken language
- speech signal
- speech synthesis
- speaker independent
- text input
- lexical features
- speaker identification
- english text
- information retrieval
- test collection
- n gram
- language model
- pattern recognition
- word level
- dialogue system
- resource allocation
- emotional speech
- text to speech
- statistical machine translation
- machine translation
- ground truth
- probabilistic model
- hidden markov models
- image segmentation