Speaker anonymization using neural audio codec language models.
Michele PanarielloFrancesco NespoliMassimiliano TodiscoNicholas W. D. EvansPublished in: CoRR (2023)
Keyphrases
- language model
- speech recognition
- audio visual
- speaker identification
- language modeling
- prosodic features
- automatic transcription
- n gram
- information retrieval
- document retrieval
- query expansion
- language modelling
- ad hoc information retrieval
- speaker verification
- multimedia
- visual information
- multi modal
- speech signal
- mixture model
- retrieval model
- probabilistic model
- automatic speech recognition
- test collection
- statistical language models
- language models for information retrieval
- speaker diarization
- visual data
- query terms
- audio features
- document ranking
- relevance model
- smoothing methods
- context sensitive
- broadcast news
- pseudo relevance feedback
- machine learning
- vector space model
- language model for information retrieval
- cross lingual
- okapi bm
- topic models
- video sequences