Inference Model with API and Integrate to LM (Language Model)
Is this model when i do inference with api, with this code :
headers = {"Authorization": f"Bearer {API_TOKEN}"}
API_URL = "https://api-inference.huggingface.co/models/ridhoalattqas/xlrs-best-lm"
def query(audio_bytes):
response = requests.request("POST", API_URL, headers=headers, data=audio_bytes)
return json.loads(response.content.decode("utf-8"))
is already connect with language_model sir ?
Hello @ridhoalattqas ,
Sorry I don't fully understand the question here - do you want to use the model with the inference API?
i want to use the model with API inference without saving the model into my local drive, but i try to inference the language_model (KenLM.arpa) is not linked with the inference model
@Narsil do you know more here by any chance? Can one use the KenLM model in the inference API?
Sorry for the long delay, new notification system really helped me see conversations where I was mentionned.
Yes currently the API should use your kenlm
model.
If you want to check, just use latest transformers and do something like pipe = pipeline(model="ridhoalattqas/xlrs-best-lm")
with kenlm
installed. Should work out of the box.