Audio-AdapterFusion: A Task-ID-free Approach for Efficient and Non-Destructive Multi-task Speech Recognition.
Hillary NgaiRohan AgrawalNeeraj GaurW. Ronny HuangParisa HaghaniPedro Moreno MengibarPublished in: CoRR (2023)
Keyphrases
- speech recognition
- multi task
- speech processing
- speaker identification
- speech recognition technology
- language model
- hidden markov models
- automatic speech recognition
- multi task learning
- audio visual speech recognition
- speech synthesis
- speech signal
- speech recognizer
- pattern recognition
- learning tasks
- signal processing
- audio visual
- noisy environments
- feature selection
- learning problems
- transfer learning
- speech recognition systems
- broadcast news
- multi modal
- multimedia