50 points | by lawik 7 days ago ago
5 comments
VAD is certainly a complex but underappreciated topic. If you like signal processing, FFTs and want to see a similar concept implemented in Python, then check out this code. It's has a fairly well-tuned VAD component built-in.
https://github.com/neuml/txtai/blob/master/src/python/txtai/...
That’s just using WebRTC-VAD, right?
No, it's using that in ensemble with a FFT and butter filters. WebRTC-VAD by itself didn't work that well.
That is some of the highest buzzword density I have seen in a while.
No mention of AI or LLM, nobody would ever fund this.
VAD is certainly a complex but underappreciated topic. If you like signal processing, FFTs and want to see a similar concept implemented in Python, then check out this code. It's has a fairly well-tuned VAD component built-in.
https://github.com/neuml/txtai/blob/master/src/python/txtai/...
That’s just using WebRTC-VAD, right?
No, it's using that in ensemble with a FFT and butter filters. WebRTC-VAD by itself didn't work that well.
That is some of the highest buzzword density I have seen in a while.
No mention of AI or LLM, nobody would ever fund this.