Low-resource language

[EklavyaFCB]

Hello,
I want to transcribe a low-resource language which is not included in the languages list in the paper.
Would this still somehow be possible ?
Just getting phonemes rather than whole words is also useful.
It's a south American language similar to Spanish.

[jongwook]

While it'd be better to fine-tune on some labeled data (e.g. https://huggingface.co/blog/fine-tune-whisper), have you tried just specifying Spanish and run Whisper? When we tried Asturian, another language adjacent to Spanish, it worked better than random.