Description:
When using the NoteCat bot to record and transcribe conversations in Spanish, the resulting transcriptions are inaccurate. It seems that the speech recognition or AI model used is primarily trained for English, which causes it to confuse many words and mix languages.
For example, common Spanish words are misspelled or replaced with English terms, making the transcripts hard to read and unreliable for meeting notes.
Suggestion:
It would be great if NoteCat could automatically detect the spoken language or allow users to manually select it (e.g., choosing “Spanish” before recording) to improve transcription accuracy.