DDSP-based Singing Vocoders: A New Subtractive-based Synthesizer and A Comprehensive Evaluation

Da-Yi Wu, Wen-Yi Hsiao, Fu-Rong Yang, Oscar D Friedman, Warren Jackson, Scott Bruzenak, Yi-Wen Liu, Yi-Hsuan Yang
2022 Zenodo  
A vocoder is a conditional audio generation model that converts acoustic features such as mel-spectrograms into waveforms. Taking inspiration from Differentiable Digital Signal Processing (DDSP), we propose a new vocoder named SawSing for singing voices. SawSing synthesizes the harmonic part of singing voices by filtering a sawtooth source signal with a linear time-variant finite impulse response filter whose coefficients are estimated from the input mel-spectrogram by a neural network. As this
more » ... approach enforces phase continuity, SawSing can generate singing voices without the phase-discontinuity glitch of many existing vocoders. Moreover, the source-filter assumption provides an inductive bias that allows SawSing to be trained on a small amount of data. Our evaluation shows that SawSing converges much faster and outperforms state-of-the-art generative adversarial network- and diffusion-based vocoders in a resource-limited scenario with only 3 training recordings and a 3-hour training time.
doi:10.5281/zenodo.7316600 fatcat:bbt2zvdlh5bkloxf2dpfrluycm