Miloš Cerňak

Phonological vocoding using artificial neural networks

2015

We investigate a vocoder based on artificial neural networks using a phonological speech representation. Speech decomposition is based on the phonological encoders, realised as neural network classifiers, that are trained for a particular language. The …

Speaker Embeddings as Individuality Proxy for Voice Stress Detection

2023 · arXiv (Cornell University)

Since the mental states of the speaker modulate speech, stress introduced by cognitive or physical loads could be detected in the voice. The existing voice stress detection benchmark has shown that the audio embeddings extracted …