Low-Latency Electrolaryngeal Speech Enhancement Based on Fastspeech2-Based Voice Conversion and Self-Supervised Speech Representation.
Kazuhiro KobayashiTomoki HayashiTomoki TodaPublished in: ICASSP (2023)
Keyphrases
- speech enhancement
- low latency
- speech signal
- noisy environments
- noise reduction
- signal to noise ratio
- single channel
- vocal tract
- speech recognition
- linear prediction
- real time
- text to speech
- highly efficient
- high speed
- virtual machine
- multi channel
- smoothing algorithm
- stream processing
- background noise
- data structure
- additive noise
- automatic speech recognition
- prediction error
- image acquisition
- high throughput