StreamSpeech: Low-Latency Neural Architecture for High-Quality on-Device Speech Synthesis.
Georgi ShopovStefan GerdjikovStoyan MihovPublished in: ICASSP (2023)
Keyphrases
- low latency
- neural architecture
- speech synthesis
- speech recognition
- text to speech
- neural network
- high speed
- feed forward
- high throughput
- real time
- virtual machine
- basis functions
- highly efficient
- multilayer perceptron
- stream processing
- multi layer perceptron
- data acquisition
- hidden markov models
- language model
- microarray
- feature selection
- information systems