openai-whisper-realtime VS plaidml

I tried running it in realtime with live audio input (kind of).

You can find the python script in this repo: https://github.com/tobiashuttinger/openai-whisper-realtime

basically agree with all of this, however I do want to highlight that there is no 'ZKML protocol plan' - the panel here are all involved in quite different projects and interested in ZKML for a variety of reasons. As one of the authors of https://github.com/plaidml/plaidml I'm not expecting any kind of standard protocol to evolve for several years; the group behind the AMA though is optimistic about the potential of ZKML and this AMA is part of the start of developing useful protocols.

It understands my Swedish attempts at English really well with the medium.en model. (Although, it gives me a funny warning: `UserWarning: medium.en is an English-only model but receipted 'English'; using English instead.`. I guess it doesn't want to be told to use English when that's all it can do.)

However, it runs very slowly. It uses the CPU on my macbook, presumably because it hasn't got a NVidia card.

Googling about that I found [plaidML](https://github.com/plaidml/plaidml) which is a project promising to run ML on many different gpu architectures. Does anyone know whether it is possible to plug them together somehow? I am not an ML researcher, and don't quite understand anything about the technical details of the domain, but I can understand and write python code in domains that I do understand, so I could do some glue work if required.

Have you tried PlaidML https://github.com/plaidml/plaidml

This doesn't answer your question, but it would be cool if we had something based on MLIR for GPU compute. From what I've read, it closes the gap between NVIDIA and other GPU vendors a lot more than pure compute shaders. e.g. ONNX-MLIR, PlaidML, and IREE.

There is a library called PlaidML which is supposed to support Keras on a wide variety of GPUs, including the Iris. But it doesn't. I get the issue reported as Issue #168, which was first reported in 2018 and is still open. That's what I mean by not well supported.

Not really: https://github.com/plaidml/plaidml/commits/plaidml-v1

There are reports on github of plaidML conking out on older CPUs with a similar "illegal instruction err.