How do I use the whisper model from Azure OpenAI API

Question

How do I use the whisper model from Azure OpenAI API

Ravi K 70

I'm not sure how to use whisper model from Azure OpenAI API. I can use curl and get it to work. I'm looking for python code to access this API.

curl https://WWHISPER_ENDPOINT/openai/deployments/whisper/audio/transcriptions?api-version=2023-09-01-preview -H "api-key: API_KEY -H "Content-Type: multipart/form-data" -F file="@./xXX.mp3"

Accepted answer

0 additional answers

Your answer

Answer 1

Saurabh Sharma 23,846 Microsoft Employee Moderator

Hi @Ravi K

Welcome to Microsoft Q&A! Thanks for posting the question.

You can use the below code in python to use the Whisper model. Please note that you need to install the latest version of openai (0.28.1)

import openai
import time
import os

openai.api_type = "azure"
openai.api_base = "https://{resourcename}.openai.azure.com/"
openai.api_key = os.getenv("AOAIKey")
openai.api_version = "2023-09-01-preview"

model_name = "whisper-1"
deployment_id = "whisper"
audio_language="en"

audio_test_file = "./hello.m4a"

result = openai.Audio.transcribe(
            file=open(audio_test_file, "rb"),            
            model=model_name,
            deployment_id=deployment_id
        )

print(result.text)

Please refer to the documentation for additional information.

Please let me know if you have any questions.

Thanks

Saurabh

Please 'Accept as answer' and Upvote if it helped so that it can help others in the community looking for help on similar topics.

Ravi K 70

Thank you Saurabh. I'm still getting the error below with the exact code. I double checked the deployment name and other parameters like api_key, base and version. They are all good.

 response = openai.Audio.transcribe(file=audio_file,model="whisper-1",deployment_id="ava-whisper")
  File "/opt/anaconda3/lib/python3.8/site-packages/openai/api_resources/audio.py", line 67, in transcribe
    response, _, api_key = requestor.request("post", url, files=files, params=data)
  File "/opt/anaconda3/lib/python3.8/site-packages/openai/api_requestor.py", line 298, in request
    resp, got_stream = self._interpret_response(result, stream)
  File "/opt/anaconda3/lib/python3.8/site-packages/openai/api_requestor.py", line 700, in _interpret_response
    self._interpret_response_line(
  File "/opt/anaconda3/lib/python3.8/site-packages/openai/api_requestor.py", line 765, in _interpret_response_line
    raise self.handle_error_response(
openai.error.InvalidRequestError: Resource not found

Saurabh Sharma 23,846 Reputation points Microsoft Employee Moderator

2023-10-10T14:43:50.4066667+00:00

Ravi K I am checking on this one and get back to you. Thanks
Ravi K 70 Reputation points

2023-10-10T14:46:51.1966667+00:00

Thank you
Ravi K 70 Reputation points

2023-10-10T16:49:21.5066667+00:00

Saurabh, I just needed to get the latest openai. Now the code works. Thank you so much
Saurabh Sharma 23,846 Reputation points Microsoft Employee Moderator

2023-10-10T17:17:10.4366667+00:00

Ravi K, Great, thanks for the confirmation.

Share via

How do I use the whisper model from Azure OpenAI API

0 additional answers

Your answer