Pretrained Speech Encoders and Efficient Fine-tuning Methods for Speech Translation: UPC at IWSLT 2022.
Ioannis TsiamasGerard I. GállegoCarlos EscolanoJosé A. R. FonollosaMarta R. Costa-jussàPublished in: IWSLT@ACL (2022)
Keyphrases
- fine tuning
- computationally expensive
- speech recognition
- preprocessing
- computationally intensive
- viable alternative
- empirical studies
- complexity analysis
- machine learning methods
- automatic speech recognition
- recognition engine
- neural network
- emotion recognition
- noisy environments
- speech signal
- benchmark datasets
- significant improvement
- video sequences
- bayesian networks
- search engine
- learning algorithm