Dual Parameter-Efficient Fine-Tuning for Speaker Representation Via Speaker Prompt Tuning and Adapters.
Zhe LiMan-Wai MakHelen Mei-Ling MengPublished in: ICASSP (2024)
Keyphrases
- fine tuning
- fine tune
- speech recognition
- viable alternative
- speaker verification
- fine tuned
- input output
- speaker diarization
- audio visual
- automatic speech recognition
- speaker identification
- speaker recognition
- objective function
- parameter settings
- synthesized speech
- cost effective
- computationally efficient
- general purpose
- data structure