Can I transcribe non-English podcasts?

Yes — but language support depends on which transcription provider you choose. Benecaster works with four providers, each with different multilingual coverage:

OpenAI Whisper is the most capable choice for non-English content. It auto-detects the language from the audio itself, so a Spanish or Portuguese podcast transcribes correctly without any additional setup on your end. AssemblyAI and Deepgram have broad language coverage, but their accuracy is highest for English and a handful of major European languages — check each provider’s documentation for specifics on your language.

Benecaster currently passes audio to your chosen provider without a language hint — the provider handles detection automatically.

If your episodes switch between languages mid-episode, keep the language setting on Auto-detect. Forcing a specific language when the audio code-switches typically makes results worse — the provider will try to transcribe everything in the forced language. Of the four providers, OpenAI Whisper handles mid-episode language switching most reliably; the others tend to commit to the dominant language detected at the start.