Enhanced Direct Speech-to-Speech Translation Using Self-supervised Pre-training and Data Augmentation.
Sravya PopuriPeng-Jen ChenChanghan WangJuan PinoYossi AdiJiatao GuWei-Ning HsuAnn LeePublished in: CoRR (2022)
Keyphrases
- data sets
- data collection
- statistical analysis
- high quality
- data processing
- training data
- speech signal
- data points
- image data
- data mining techniques
- speech recognition
- labeled data
- audio stream
- raw data
- missing data
- learning algorithm
- database
- data sources
- data structure
- databases
- training samples
- sensor data
- experimental data
- test data
- knowledge discovery