Hi , I am using custom trained model for speech to text for spanish language , how can I get the transcription without diacritics. I tried using custom display formatting using rewrite . But it seems it does not work. Any help is really appreciated

Question

These are the examples I tried in custom display formatting using #rewrite

- sí	si
- qué	que
- gustaría	gustaria
- ¿Puedo	puedo
- balsámico	balsamico

Answer 1

Thanks for the question, To remove diacritics from the transcription in a custom trained model for speech to text, you can post-process the output of your model. Here’s a general approach: Transcribe the Speech: Use your custom trained model to transcribe the speech to text.

Remove Diacritics: After you get the transcription, you can use a Python library like unidecode to remove diacritics from the Spanish text.