How can I cancel transcription mid-way while using Whisper for long audio?

[C-Jaewon]

Hi,

I'm using OpenAI Whisper to transcribe long audio files (sometimes over 30 minutes). I'm looking for a way to cancel the transcription process partway through, especially in scenarios where the user wants to stop the process or skip the rest of the file.

Currently, I’m calling model.transcribe() in Python. However, I don’t see any built-in option to interrupt the process once it has started. I would like to ask:

❓ My questions:
Is there any supported or recommended way to cancel an ongoing transcription mid-way using the standard Whisper API?

If not supported directly, is there a clean workaround—such as using multiprocessing, timeouts, or chunking manually—to safely interrupt transcription?

I’d appreciate any suggestions, examples, or workarounds others have used.

Thanks in advance!

[phineas-pta]

u should look for asynchronous programming in python, for example how to interrupt a long running coroutines