Sorry, you need to enable JavaScript to visit this website.

Differential DSP Vocoder - ICASSP 2024

DOI:
10.60864/a0ff-ke67
Citation Author(s):
Submitted by:
Prabhav Agrawal
Last updated:
6 June 2024 - 10:23am
Document Type:
Presentation Slides
Document Year:
2024
Presenters:
Thilo Koehler, Prabhav Agrawal
Paper Code:
SLP-L18.1
 

Neural vocoders model the raw audio waveform and synthesize highquality audio, but even the highly efficient ones, like MB-MelGAN
and LPCNet, fail to run real-time on a low-end device like a smartglass. A pure digital signal processing (DSP) based vocoder can
be implemented via lightweight fast Fourier transforms (FFT), and
therefore, is a magnitude faster than any neural vocoder. A DSP
vocoder often gets a lower audio quality due to consuming oversmoothed acoustic model predictions of approximate representations
for the vocal tract. In this paper, we propose an ultra-lightweight differential DSP (DDSP) vocoder that uses a jointly optimized acoustic
model with a DSP vocoder, and learns without an extracted spectral feature for the vocal tract. The model achieves audio quality
comparable to neural vocoders with a high average MOS of 4.36
while being efficient as a DSP vocoder. Our C++ implementation,
without any hardware-specific optimization, is at 15 MFLOPS, surpasses MB-MelGAN by 340 times in terms of FLOPS, and achieves
a vocoder-only RTF of 0.003 and overall RTF of 0.044 while running
single-threaded on a 2GHz Intel Xeon CPU.

up
0 users have voted: