Voice Conversion Based Speaker Normalization for Acoustic Unit Discovery.
Thomas GlarnerJanek EbbersReinhold Häb-UmbachPublished in: CoRR (2021)
Keyphrases
- speech sounds
- prosodic features
- mel frequency cepstral coefficients
- speaker verification
- text to speech
- acoustic features
- automatic speech recognition
- speaker recognition
- speaker identification
- speech recognition
- speech signal
- vocal tract
- speech synthesis
- emotion recognition
- synthesized speech
- data sets
- scientific discovery
- preprocessing
- feature set
- knowledge discovery
- speaker diarization
- source localization
- acoustic models
- data mining
- principal component analysis
- normalization method
- feature vectors
- feature space
- emotional speech
- information retrieval