FusedF0: Improving DNN-based F0 Estimation by Fusion of Summary-Correlograms and Raw Waveform Representations of Speech Signals.
Eray ErenLee Ngee TanAbeer AlwanPublished in: INTERSPEECH (2023)
Keyphrases
- speech signal
- speech recognition
- fundamental frequency
- automatic speech recognition
- automatic speech recognition systems
- sound signals
- vocal tract
- noisy environments
- additive noise
- hidden markov models
- noisy images
- blind source separation
- non stationary
- spectral analysis
- linear prediction
- adaptive filtering
- background noise
- neural network
- sound source
- visual features
- markov random field
- edge detection
- speech sounds