Closed malradhi closed 7 years ago
source filter model, but with the '+' happening in log domain.
just to precise: as a consequence, it is neither the traditional source filter model nor the sinusoidal model. It would be more correct to call it a convolutional model (the noise component being convolved with the deterministic component and not added as in the traditional source filter model)
Many thanks for your helpful explanation.
You're welcome!
Dear all,
Does this type of vocoder considered as a source filter model or sinusoidal model?
Thank you very much in advance