close text postprocessing

#57
by Weiweier - opened

I am using the ASR model on your platform, and I would like to disable text normalization in the output. Specifically, I want the model to output the recognized English words exactly as spoken, without converting them into numbers or symbols (for example, I want "zero point four percent" instead of "0.4%").
Is there a way to turn off or customize the text normalization/postprocessing so that the output remains in the original spoken form? Any guidance or suggestions would be greatly appreciated.

Sign up or log in to comment