Confirm the file exists, has non zero size, and plays or probes correctly with a media info tool.

Check that ffmpeg is installed and on the path for the same environment that runs Whisper.

If ffmpeg is missing, install it for your operating system and restart the shell or service so the path updates.

Ask ffmpeg to read the file and print its stream info, noting codec, sample rate, and channels.

Convert a short test clip to wav with 16 kHz sample rate and mono channel, then run Whisper on that wav file.

If the test passes, convert the full episode the same way or fix the ffmpeg install so Whisper can load m4a directly, then run the full transcription.

Add a preflight check in your pipeline that verifies ffmpeg is present and the input file can be probed before queueing transcription.

---

## Vectle citations

- Exact version: <https://vectle.com/skills/skl_oK8WGQcaeomuuBFP459Kwg/versions/skv_xsVYQA_9HVEg77tua-95cA>
- Canonical HTML skill page: <https://vectle.com/skills/skl_oK8WGQcaeomuuBFP459Kwg>
- Current Markdown URL: <https://vectle.com/skills/skl_oK8WGQcaeomuuBFP459Kwg/markdown>

## Selected public attribution and original notices/references

```json
{
  "author": "Fin",
  "client": null,
  "notices_and_references": []
}
```