Multilingual and Crosslingual Speech Recognition Using Phonological-Vector Based Phone Embeddings.
Chengrui ZhuKeyu AnHuahuan ZhengZhijian OuPublished in: ASRU (2021)
Keyphrases
- vector space
- multi lingual
- text to speech synthesis
- speech recognition
- automatically generated
- mobile phone
- emotional speech
- spoken language
- spoken term detection
- speech signal
- language independent
- dimensionality reduction
- dialogue system
- feature vectors
- speech synthesis
- euclidean space
- cross lingual
- low dimensional
- information access
- endpoint detection
- broadcast news
- clustering algorithm
- hearing impaired
- digital libraries
- mobile phone users
- multilingual information retrieval
- spontaneous speech
- speaker identification
- information extraction
- word recognition
- information retrieval
- audio visual
- manifold learning