You probably still better use inference on ANE (Apple Neural Engine) via CoreML rather than Metal - speed will be either similar or even faster on non-pro macbooks or iphones and power consumption significantly better. Metal or even MLX format doesn't have to be the fastest and the only way to access ANE is via CoreML.<p>Can use this library:<p><a href="https://github.com/FluidInference/FluidAudio" rel="nofollow">https://github.com/FluidInference/FluidAudio</a>
I built a C++ inference engine for NVIDIA's Parakeet speech recognition models using Axiom(<a href="https://github.com/Frikallo/axiom" rel="nofollow">https://github.com/Frikallo/axiom</a>) my tensor library.<p>What it does:
- Runs 7 model families: offline transcription (CTC, RNNT, TDT, TDT-CTC), streaming (EOU,
Nemotron), and speaker diarization (Sortformer)
- Word-level timestamps
- Streaming transcription from microphone input
- Speaker diarization detecting up to 4 speakers
Related:<p><a href="https://github.com/antirez/qwen-asr" rel="nofollow">https://github.com/antirez/qwen-asr</a><p><a href="https://github.com/antirez/voxtral.c" rel="nofollow">https://github.com/antirez/voxtral.c</a><p>Qwen-asr can easily transcribe live radio (see README) in any random laptop. It looks like we are going to see really cool things on local inference, now that automatic programming makes a lot simpler to create solid pipelines for new models in C, C++, Rust, ..., in a matter of hours.
Your voxtral.c work was a big motivator for me. I built a macOS menu bar dictation app (<a href="https://github.com/T0mSIlver/localvoxtral" rel="nofollow">https://github.com/T0mSIlver/localvoxtral</a>) around Voxtral Realtime, currently using a voxmlx fork with an OpenAI Realtime WebSocket server I added on top.<p>The thing that sold me on Voxtral Realtime over Whisper-based models for dictation is the causal encoder. Text streaming in as you speak rather than appearing after you stop is a fundamentally different UX. On M1 Pro with a 4-bit quant through voxmlx it feels responsive enough for natural dictation, though I haven't done proper latency benchmarks yet.<p>Integrating voxtral.c as a backend is on my roadmap, compiling to a single native binary makes it much easier to bundle into a macOS app than a Python-based backend.
Which is why long term current programming languages will eventually become less relevant in the whole programming stack, as in get the computer to automate tasks, regardless how.
Assuming RAM prices will not make it totally unaffordable. Current situation is atrocious and big infrastructure corps seem to love it, they do not want independent computing. Alternatively they might build specialized branded hardware which people could only use for what corps allow them to do for nice monthly fee.<p>Another problem is too much abstraction on input spec level. The other day I asked Claude to generate few classes. When reviewing the code I noticed it doing full scan for ranges on one giant set. This would bring my backend to a halt. After pointing it out to Claude it had smartened up to start with lower_bound() call. When there are no people to notice such things what do you think we are going to have?
> Alternatively they might build specialized branded hardware which people could only use for what corps allow them to do for nice monthly fee.<p>That's why I'm still holding on to a bulky Core 2 Duo Management Engine-free Fujitsu workstation, for when personal computing finally goes underground again.
Agreed, in regards to prices, it appears to be the new gold, lets see how this gets sorted out, with NPUs, FPGAs, analog (Cerebas),...<p>Now the abstraction I am with you on that, I foresee a more formal way to give specifications, but more suitable for natural language as input, or even proper mathematics, than the languages we have been using thus far.<p>Naturally we aren't there yet.
Off topic but if anyone is looking for a nice web-GUI frontend for a locally-hosted transcription engine, Scriberr is nice<p><a href="https://github.com/rishikanthc/Scriberr" rel="nofollow">https://github.com/rishikanthc/Scriberr</a>
I've been using handy with parakeet on both Windows and mac, and have been very impressed.<p>Hoe does this compare?
I was also impressed with Handy.<p>I played around with it this week, and when you enable advanced mode and add a post-transcription AI model to point to your own server which mimics a minimal ChatGPT-compatible behavior, then you can use it to modify the output, even return an empty string if you noticed that the transcript was more targeted to do other stuff ("turn the lights on"), if you then return an empty string, it won't inject keypresses.<p>So one gets the best for both worlds: transcription for dictation and transcription to trigger events.<p>If I now only could let it listen constantly and react to voice, so that no push to talk is active, that would be nice.<p>Maybe this project here could be used for that.<p>Also, this seems to support streaming transcription.
Handy only supports microphone input not files afaik
Is there anything truly low latency(sub 100ms)? Speech recognition is so cool but I want it to be low latency.
Parakeet does streaming I think, so if you throw enough compute at it, it should be. The closest competitor is whisper v3 which is relatively slow, maybe Voxtral but it's still very new.
[dead]