From Raw Speech to Fixed Representations: A Comprehensive Evaluation of Speech Embedding Techniques.
Dejan PorjazovskiTamás GrószMikko KurimoPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2024)
Keyphrases
- comprehensive evaluation
- speech recognition
- speech signal
- automatic speech recognition
- audio visual
- endpoint detection
- information retrieval
- speech recognizer
- speech synthesis
- text to speech
- language acquisition
- dialogue system
- speaker recognition
- vector space
- systematic evaluation
- recognition engine
- spoken language
- speaker verification
- semantic network
- speech processing
- gaussian mixture model
- genetic algorithm