Groq provides whisper models. I see...
https://console.groq.com/docs/speech-text

Yes, exactly. That's why i need the base_url parameter. To use groq i need to put their url there and their api key.

Hey, could you merge the PR #795 ?

Thanks the PR, but I want to use groq library.
Draft: a099938

In Groq documentation, groq library is used.
From my viewpoint, openai library with base_url seems kind of hack for workaround.

@amosdevstudio Try SpeechRecognition 3.12.0, which supports Groq.

python -m pip install SpeechRecognition==3.12.0

>>> # NOTICE: export GROQ_API_KEY before launching Python interpreter
>>> import speech_recognition as sr
>>> r = sr.Recognizer()
>>> with sr.AudioFile("english.wav") as source:
...   audio = r.record(source)
>>>
>>> r.recognize_groq(audio)
' One, two, three.'

Great! I love to see this.