parser.add_argument("--task", type=str, default="transcribe", choices=["transcribe", "translate"], help="whether to perform X->X speech recognition ('transcribe') or X->English translation ('translate')")
    parser.add_argument("--language", type=str, default=None, choices=sorted(LANGUAGES.keys()) + sorted([k.title() for k in TO_LANGUAGE_CODE.keys()]), help="language spoken in the audio, specify None to perform language detection")

found this in transcribe.py of talkscribers whisper repo. there is no default language value and the two modes are transcribe and translate. the fact that both on the dashboard and via web sdk, it will work temporarily then automatically translate to english makes me think either vapi has a default value set for the language param (english) or that parameter is just ignored. I have the mode parameter set to transcribe yet it still translates every language into english transcripts, even with the language parameter removed. the mode of transcribe shoudl preserve the language of whats spoken rather than always translate it to english before logging it in the transcription