mfcc-rust VS openWakeWord

hey, quick question. do you mind if I use your stft function in the speech preprocessing library I've been working on? we've been trying to add support for doing mel spectrograms to build a runner for openwakeword, but progress is pretty slow because I've been soloing something I really don't have the right background for(I've never directly studied or worked with signal processing)

So I've been working on an speech preprocessing library for rust for a minute now. The tl;dr is it contains functions related to preprocessing audio for use in speech-centric models. I've run into a situation where I'm running against a few conflicting design constraints and trying to figure out the best way to handle state for a set of functions.

we could always use contributors for mfcc-rust/speechsauce(we're planning on changing the name as it's not limited to just computing mfcc's).

The code has comments and in the GitHub issues is written what's missing, the goal is to mimic functionality from NumPy not present in NDArray (a library for maths operations). You can find the code here: https://github.com/secretsauceai/mfcc-rust. Take a look, if you feel that you can give a hand, that would be amazing!!! Let's make a new exciting voice future!!!

This release represents the first phase in the wakeword project, we are working on a Rust wakeword engine based on Precise and a SpeechPy MFCC port in Rust so that user's can run the wakeword easily on their phone and other devices. It is hard to believe that there aren't any current good solutions to running a modern FOSS wakeword engine on a phone in real time. We want to change that and allow everyone access to this technology, with their own wakeword of choice.

https://github.com/dscripka/openWakeWord

Balancing wake reliability vs false wake activation is a tricky balance. OWW is decent but could certainly be better.

It's used with Home Assistant now so I expect the training data and implementation overall to get significantly better fairly soon.

There's also OpenWakeWord[0]. The models are readily available in tflite and ONNX formats and are impressively "light" in terms of compute requirements and performance.

It should be possible.

[0] - https://github.com/dscripka/openWakeWord

hey, quick question. do you mind if I use your stft function in the speech preprocessing library I've been working on? we've been trying to add support for doing mel spectrograms to build a runner for openwakeword, but progress is pretty slow because I've been soloing something I really don't have the right background for(I've never directly studied or worked with signal processing)

potentially build another runner for open wakeword

here's what's on the pipeline next: - finish mel-spectrogram implementation - publish initial version on crates - move python caching rust side - finish implementing in the precise rust port - potentially build another runner for (open wakeword)[https://github.com/dscripka/openWakeWord] - build an android app that supports user-defined wakewords and has some popular defaults to load. ps not a voice assistant, just the thing that activates the voice assistant.