I-vector Based Within Speaker Voice Quality Identification on connected speech.
Chuyao FengEva van LeerMackenzie Lee CurtisDavid V. AndersonPublished in: CoRR (2021)
Keyphrases
- speech recognition
- speech sounds
- synthesized speech
- prosodic features
- audio visual
- automatic speech recognition
- speaker recognition
- speech quality
- high quality
- text to speech
- speaker verification
- speech signal
- speech synthesis
- emotion recognition
- vocal tract
- speaker identification
- speaker dependent
- quality measures
- mel frequency cepstral coefficients
- speech recognition errors
- voice activity detection
- broadcast news
- linear prediction
- noisy environments
- feature vectors
- speaker diarization
- recognition engine
- speaker independent
- quality assessment
- similarity measure
- multimedia