3 ms·
I’m building SPARK (Signal Processing Algorithms, Routines, and Kernels), an open-source library of modular, efficient DSP components for low-power embedded aud
by ashafq 1y ago
I’m building SPARK (Signal Processing Algorithms, Routines, and Kernels), an open-source library of modular, efficient DSP components for low-power embedded audio systems.
The goal is to make it straightforward to design and deploy small, composable audio graphs that fit on MCUs and similar hardware. The project is in its infancy, so there’s plenty of room for experimentation and contributions.
https://github.com/Colahall/SPARK https://github.com/Colahall/SPARK
- kragen 1y agoNice! Unrelated to the other open-source SPARK for designing and deploying small, composable data processing graphs with a Python API, I guess? Are you thinking about supporting deployment on FPGAs like the iCE40 line?
- ashafq 1y agoCurrently FPGAs are out of scope. But, would love to implement that use case as well in future.
- kragen 1y agoI've been clumsily exploring the space of what kind of DSP you can do without multipliers, because (in hardware) multiplication is one or two orders of magnitude more expensive than addition; see https://nbviewer.org/url/canonical.org/~kragen/sw/dev3/sparse-gabor.ipynb https://nbviewer.org/url/canonical.org/~kragen/sw/dev3/spars... for one example.