Katukagloria/fast

samples/fast-transcription/python_client/main.py

@@ -0,0 +1,191 @@
+# Copyright (c) Microsoft. All rights reserved.
+# Licensed under the MIT license. See LICENSE.md file in the project root for full license information.
+
+import requests
+import os
+import logging
+import json
+import warnings
+warnings.filterwarnings("ignore")
+from azure.storage.blob import BlobServiceClient
+from azure.identity import DefaultAzureCredential
+
+
+# Your Credentials and secrets - You should create a .env file and store your private credentials
+SUBSCRIPTION_KEY = "YourSubscriptionKey"
+SERVICE_REGION = "YourServiceRegion"
+STORAGE_ACCOUNT_NAME = 'YourStorageAccountName'
+INPUT_CONTAINER =  'YourBlobContainerName' #This is the input container with audio recordings 
+OUTPUT_CONTAINER = 'YourOutputBlobContainer'
+RECORDINGS_BLOB_URL = f"https://{STORAGE_ACCOUNT_NAME}.blob.core.windows.net" #"Your storage blob url"
+ENPOINT_URL = f"https://{SERVICE_REGION}.api.cognitive.microsoft.com/speechtotext/transcriptions:transcribe?api-version=2024-05-15-preview"
+
+# Fast transcription description properties
+LOCALE = ["en-US"]
+PROFANITYFILTER = "Masked"
+DIARIZATIONSETTINGS =  {"minSpeakers": 1, "maxSpeakers": 4}
+
+def get_transcription_properties():
+    #transcription "definition" payload 
+    transcription_definition = {
+        "profanityFilterMode": PROFANITYFILTER,
+        "channels": [0], #Change to one channel for Mono-speaker diarization
+        "diarizationSettings": DIARIZATIONSETTINGS,  # Optional
+         }
+
+    #transcription properties 
+    transcription_properties = {
+            "locales": LOCALE,
+            "properties": transcription_definition
+        }
+    return transcription_properties
+
+def transcribe_from_local(audio_file_path, api_key):
+    """
+    Transcribe a local file using the settings specified in `properties`
+    using the base model for the specified locale.
+
+    files (dict): Dictionary of Multi-part form data to send.
+    audio_file_path (str): Path to the file to be uploaded.
+    """
+    #Note: Content-Type should not be set manually as requests will handle it 
+    headers = {
+      "Ocp-Apim-Subscription-Key": api_key
+    }
+
+    # Check if the file exists in the specified file path
+    if not os.path.exists(audio_file_path):
+        print("File does not exist at the specified path")
+        return
+    else:    
+        #transcription properties payload 
+        transcription_properties = get_transcription_properties()
+
+        #Multipart transcription files payload 
+        files = {
+        'audio': ('test_call_audio.wav', open(audio_file_path, 'rb'), 'audio/wav'),
+        # 'audio': ('test_call_audio2.wav', open(audio_file_path2, 'rb'), 'audio/wav'), #add multiple audio files
+        'definition': (None, json.dumps(transcription_properties), 'application/json')
+        }
+
+        endpoint_url = ENPOINT_URL
+
+        logging.info("Starting transcription client...")
+
+        response = requests.post(endpoint_url, headers=headers, files=files)
+        print(response.status_code)
+        response_json = json.dumps(response.json(), indent=4)
+        # print(response.json())
+
+        if response.status_code == 200:
+            response_json = response.json()
+            # Write response to a JSON file
+            with open("transcription_response.json", "w") as json_file:
+                json.dump(response_json, json_file, indent=4)
+
+
+def get_blob_container_url(storage_account_url, container_name):
+    """
+    Retrieve the URL of a Blob Container within Azure Blob Storage.
+
+    storage_account_url (str): The base URL of the Azure Storage Account.
+    container_name (str): The name of the container whose URL is to be retrieved.
+    Returns:
+    str: The URL of the specified Blob Container.
+    """
+    try:
+        # Create the BlobServiceClient object
+        blob_service_client = BlobServiceClient(account_url=storage_account_url)
+
+        # Get the container client using the container name
+        container_client = blob_service_client.get_container_client(container_name)
+
+        # Check if the container exists
+        if container_client.exists():
+            # Return the Blob Container URL
+            container_url = container_client.url
+            print(f"Container URL: {container_url}")
+            return container_url
+        else:
+            print("Container not found.")
+            return None
+    except Exception as e:
+        print(f"An error occurred: {e}")
+        return None
+
+
+
+def transcribe_from_blob(blob_storage_url, api_key):
+    """
+    Transcribe all files in the container located at blob_storage_url using the settings specified in `properties`.
+    """  
+    storage_account_url = blob_storage_url
+    container_name = INPUT_CONTAINER #'YourInputContainerName'
+
+
+    cred = DefaultAzureCredential()
+    blob_service_client = BlobServiceClient(storage_account_url,credential=cred)
+
+    #use the client to connect to the container
+    container_client = blob_service_client.get_container_client(container_name)
+
+    #get a list of all blob files in the container
+    blob_list = []
+    for blob_i in container_client.list_blobs():
+        blob_list.append(blob_i.name)
+        blob_client = container_client.get_blob_client(blob_i.name)
+        stream = blob_client.download_blob()
+        data = stream.readall()    
+
+    #Note: Content-Type should not be set manually as requests will handle it
+    headers = {
+      "Ocp-Apim-Subscription-Key": api_key
+    }
+
+    #transcription properties payload 
+    transcription_properties = get_transcription_properties()
+    #Multipart transcription files payload 
+    files = {
+        'audio': ('test_call_audio.wav', data, 'audio/wav'),
+        'definition': (None, json.dumps(transcription_properties), 'application/json')
+        }
+
+
+    endpoint_url = ENPOINT_URL
+
+    logging.info("Starting transcription client...")
+
+    response = requests.post(endpoint_url, headers=headers, files=files)
+    print(response.status_code)
+    response_json = json.dumps(response.json(), indent=4)
+    # print(response_json)
+
+    # Create the container if it does not already exist
+    try:
+        output_container_name = OUTPUT_CONTAINER
+        container_client = blob_service_client.get_container_client(output_container_name)
+        container_client.create_container()  # This will do nothing if container already exists
+        print(f"Container {output_container_name} created or already exists.")
+    except Exception as e:
+        print(f"Could not create container: {str(e)}")
+
+
+    # Get a client for the target blob and upload the JSON response
+    if response.status_code == 200:
+        output_blob_name = f"{OUTPUT_CONTAINER}_transcriptions"
+        blob_service_client = container_client.get_blob_client(output_blob_name)
+        blob_service_client.upload_blob(response_json, overwrite=True)
+        print(f"Response JSON has been uploaded to {output_blob_name} in {output_container_name}.")
+    else:
+        print(f"Failed to get valid response from API: {response.status_code}, {response.text}")
+
+
+if __name__ == "__main__":  
+
+    #Test 1: Using audio file from your local storage
+    audio_file_path = "test_call_audio.wav" #"Your_audio_file_path"
+    transcribe_from_local(audio_file_path, SUBSCRIPTION_KEY)
+
+    ## Test 2: Using Azure blob storage container
+    # cred = DefaultAzureCredential() 
+    # transcribe_from_blob(RECORDINGS_BLOB_URL, SUBSCRIPTION_KEY)

samples/fast-transcription/python_client/readme.md

@@ -0,0 +1,30 @@
+# How to use the Speech Services Fast Transcription API from Python
+
+## Prerequisites
+
+1. An Azure AI Speech resource in one of the regions where the fast transcription API is available. The supported regions are: Central India, East US, Southeast Asia, and West Europe. For more information and region identifier, see [Speech service regions](https://learn.microsoft.com/en-us/azure/ai-services/speech-service/regions)
+1. An audio file (less than 2 hours long and less than 200 MB in size) in one of the formats and codecs supported by the batch transcription API. For more information about supported audio formats, see [Supported audio formats](https://learn.microsoft.com/en-us/azure/ai-services/speech-service/batch-transcription-audio-data?tabs=portal#supported-audio-formats-and-codecs)
+1. For additional explanation, see [fast transcription documentation](https://learn.microsoft.com/en-us/azure/ai-services/speech-service/fast-transcription-create)
+
+## Install other dependencies
+
+The sample uses the `requests` library. You can install it with the command
+
+```bash
+pip install requests
+```
+
+## Run the sample code
+
+The sample code itself is [main.py](python_client/main.py) and can be run using Python 3.8 or higher.
+You will need to adapt the following information to run the sample:
+
+1. Your AI Services subscription key and region.
+
+    Some notes:
+    - You can get the subscription key from the "Keys and Endpoint" tab on your Speech resource in the Azure Portal.
+
+1. The audio fileThe URI of an audio recording in blob storage. Please refer to the [Azure Storage documentation](https://docs.microsoft.com//rest/api/storageservices/authorize-requests-to-azure-storage) on information on how to authorize accesses against blob storage.
+
+You can use a development environment like Visual Studio Code to edit, debug, and execute the sample.
+