Prosodic-Enhanced Siamese Convolutional Neural Networks for Cross-Device Text-Independent Speaker Verification.
Sobhan SoleymaniAli DaboueiSeyed Mehdi IranmaneshHadi KazemiJeremy M. DawsonNasser M. NasrabadiPublished in: BTAS (2018)
Keyphrases
- speaker verification
- prosodic features
- convolutional neural networks
- text to speech
- text to speech synthesis
- noisy environments
- speaker recognition
- language identification
- emotion recognition
- information retrieval
- speech synthesis
- audio visual
- convolutional network
- speech recognition
- text mining
- spontaneous speech
- using artificial neural networks
- multiscale
- automatic speech recognition