An overlap-add technique based on waveform similarity (WSOLA) for high quality time-scale modification of speech.
Werner VerhelstMarc RoelandsPublished in: ICASSP (2) (1993)
Keyphrases
- high quality
- fundamental frequency
- similarity measure
- image quality
- similarity measurement
- low quality
- speech recognition
- similarity function
- speech signal
- ground truth
- audio visual
- neural network
- recognition engine
- text to speech
- speaker recognition
- automatic speech recognition
- higher quality
- super resolution
- semantic similarity
- user defined
- scale space
- depth map
- distance measure
- small scale
- data sets
- wordnet