1. Request access through the model page on Hugging Face and wait for approval from the model authors.
2. Create a Custom keys workspace connection in Microsoft Foundry or Azure Machine Learning named HuggingFaceTokenConnection with the key HF_TOKEN and value your Hugging Face read or fine-grained token (marked as secret).
3. Create the Managed Online Endpoint with the property enforce_access_to_default_secret_stores set to enabled so it can access the secret connection value.
4. Once access is approved, the connection is configured, and the endpoint is created with read access to the token, you can deploy and use the model in Microsoft Foundry or Azure Machine Learning.

• Original Model Card
• automatic-speech-recognition Task on Hugging Face

Send Request

curl <AZUREML_ENDPOINT_URL> \
    -X POST \
    -H "Authorization: Bearer <AZUREML_TOKEN>" \
    -H "Content-Type: application/json" \
    -d '{"inputs":"https://github.com/ggml-org/whisper.cpp/raw/b0a11594aec50892a02cd8d129eee2dfe93a8bb8/samples/jfk.wav","parameters":{"return_timestamps":true}}'

Supported Parameters

• inputs (string): The input audio data as a base64-encoded string. If no parameters are provided, you can also provide the audio data as a raw bytes payload.
• parameters (object):
  • return_timestamps (boolean): Whether to output corresponding timestamps with the generated text.
  • generation_parameters (object):
    • temperature (float): The value used to modulate the next token probabilities.
    • top_k (integer): The number of highest probability vocabulary tokens to keep for top-k-filtering.
    • top_p (float): If set to float < 1, only the smallest set of most probable tokens with probabilities that add up to top_p or higher are kept for generation.
    • typical_p (float): Local typicality measures how similar the conditional probability of predicting a target token next is to the expected conditional probability of predicting a random token next, given the partial text already generated. If set to float < 1, the smallest set of the most locally typical tokens with probabilities that add up to typical_p or higher are kept for generation. See this paper for more details.
    • epsilon_cutoff (float): If set to float strictly between 0 and 1, only tokens with a conditional probability greater than epsilon_cutoff will be sampled. In the paper, suggested values range from 3e-4 to 9e-4, depending on the size of the model. See Truncation Sampling as Language Model Desmoothing for more details.
    • eta_cutoff (float): Eta sampling is a hybrid of locally typical sampling and epsilon sampling. If set to float strictly between 0 and 1, a token is only considered if it is greater than either eta_cutoff or sqrt(eta_cutoff) * exp(-entropy(softmax(next_token_logits))). The latter term is intuitively the expected next token probability, scaled by sqrt(eta_cutoff). In the paper, suggested values range from 3e-4 to 2e-3, depending on the size of the model. See Truncation Sampling as Language Model Desmoothing for more details.
    • max_length (integer): The maximum length (in tokens) of the generated text, including the input.
    • max_new_tokens (integer): The maximum number of tokens to generate. Takes precedence over max_length.
    • min_length (integer): The minimum length (in tokens) of the generated text, including the input.
    • min_new_tokens (integer): The minimum number of tokens to generate. Takes precedence over min_length.
    • do_sample (boolean): Whether to use sampling instead of greedy decoding when generating new tokens.
    • early_stopping (enum): Possible values: never, true, false.
    • num_beams (integer): Number of beams to use for beam search.
    • num_beam_groups (integer): Number of groups to divide num_beams into in order to ensure diversity among different groups of beams. See this paper for more details.
    • penalty_alpha (float): The value balances the model confidence and the degeneration penalty in contrastive search decoding.
    • use_cache (boolean): Whether the model should use the past last key/values attentions to speed up decoding.