ASR = automatic speech recognition. I think this is the same as STT, speech to text?
It seems to want an NVidia GPU. I didn't look closely to find out if it can run usably without one. For me, usable might mean 10% of realtime, i.e. an overnight run on my old x86 box for 1 hour of transcription. Of course faster is always nice. YMMV.
Almost all ML models come with reference python code that will run on most architectures.
If you want an implementation optimized for a certain architecture the cost is between 10-30USD of opus tokens and is oneshottable: ("implement Nvidia nemotron 3.5 transcribe for the intel npu on this machine. Use the hf transformers implementation as reference. The finished artifact takes pcm on stdin and utf8 chunks on stdout".)
You can substitute your old intel CPU by modifying the prompt by replacing "intel npu" with "the target cou on this machine ... Your task is complete when the model decode performance is 80% of the memory read speed of this machine"
> Transcribe is built for recorded audio, not live streaming.
Bit unfortunate! Nemotron Streaming in Handy has basically solved dictation for me. It recognizes German mixed with English terms and brand names really well.
There are apps on the Google Play Store where you can run on-device models or connect to OpenRouter where they are run for you. Im using Qwen 3 ASR locally on my android, it's the second most accurate open model after the one in this post.
I played around with Nemotron 3 recently. It's OK, but makes an error every few sentences, at least on the data I worked with (standard vocab). I'm sure we'll get there, but I was pleasantly surprised by how I can recognize words better than it can.
testing it against parakeet right now for ASR, it's slower, heavier, and has a higher error rate. but i'm doing a pass with the embedded translation to see if that's any good, it might be worth it.
e: also i'm testing it with a Q4 quant that might not be full accuracy
tried running it against a full anime episode, both vocal isolated and raw. it got confused by english words in the theme song and failed spectacularly three times
I have a few thousand audio interviews as MP3, around an hour each and with specialized vocabulary. Typically there are no more than 2 to 3 speakers. Audio quality ranges from mediocre to fair to good.
I was thinking of creating a local app that calls OpenAI’s VTT model. In my most recent evaluation, the OpenAI model was best for my use case. But is that indeed the case?
It seems to want an NVidia GPU. I didn't look closely to find out if it can run usably without one. For me, usable might mean 10% of realtime, i.e. an overnight run on my old x86 box for 1 hour of transcription. Of course faster is always nice. YMMV.
If you want an implementation optimized for a certain architecture the cost is between 10-30USD of opus tokens and is oneshottable: ("implement Nvidia nemotron 3.5 transcribe for the intel npu on this machine. Use the hf transformers implementation as reference. The finished artifact takes pcm on stdin and utf8 chunks on stdout".)
You can substitute your old intel CPU by modifying the prompt by replacing "intel npu" with "the target cou on this machine ... Your task is complete when the model decode performance is 80% of the memory read speed of this machine"
Bit unfortunate! Nemotron Streaming in Handy has basically solved dictation for me. It recognizes German mixed with English terms and brand names really well.
Select nemotron in the model selection
Also, there's no mention of Nemotron in that handy. Is that the same Handy GGP was talking about?
I’m still sad Voxtral 3 wasn’t made open source by Mistral.
e: also i'm testing it with a Q4 quant that might not be full accuracy
I was thinking of creating a local app that calls OpenAI’s VTT model. In my most recent evaluation, the OpenAI model was best for my use case. But is that indeed the case?