Que nous apprennent les gros corpus sur l'harmonie vocalique en français ? (What can we learn from big speech corpora about French vowel harmony?).
Giuseppina TurcoCécile FougeronNicolas AudibertPublished in: JEP-TALN-RECITAL (1) (2016)
Keyphrases
- speech signal
- speech synthesis
- vocal tract
- prosodic features
- formant frequencies
- spontaneous speech
- text corpora
- speech recognition
- fundamental frequency
- annotated corpus
- text to speech
- document corpus
- topic segmentation
- statistical machine translation
- speech recognition systems
- text corpus
- training corpus
- conversational speech
- wide coverage
- automatic speech recognition
- parallel corpus
- text data
- linguistic patterns
- text collections
- big data
- spoken language
- mono lingual
- human machine interaction
- natural language processing
- lexical features
- linear prediction
- noisy environments
- sentence pairs
- speaker independent
- news corpus
- cross lingual
- text classification
- text mining
- vowel phonemes
- hidden markov models
- specific domains
- chinese english
- parallel corpora
- linguistic features
- manually annotated
- cross language information retrieval
- named entities
- hand crafted
- speaker identification
- broadcast news
- dialogue system
- knn
- pattern recognition