Update main.py

main.py

@@ -4,6 +4,13 @@ from transformers import pipeline
 from fastapi import BackgroundTasks, File, UploadFile, FastAPI
 from fastapi.responses import RedirectResponse
 
+from azure.storage.blob import BlobServiceClient
+
+STORAGEACCOUNTURL = "https://callreviewer.blob.core.windows.net"
+STORAGEACCOUNTKEY = "vXq0X89zOaQxQmv7UBGFqqa61V0FRE6Gx1TgJvbtxZn5zLJ1ETc9aGDbbotuSoQzf5ob9QTuXlof+AStdHXOpA=="
+CONTAINERNAME = "default"
+BLOBNAME = "audio.mp3"
+
 MODEL_NAME = "openai/whisper-large-v2"
 
 device = 0 if torch.cuda.is_available() else "cpu"
@@ -19,12 +26,17 @@ all_special_ids = pipe.tokenizer.all_special_ids
 transcribe_token_id = all_special_ids[-5]
 translate_token_id = all_special_ids[-6]
 
-def transcribe(file: UploadFile):
+def transcribe():
     try:
         print("For processing...")
+        blob_service_client_instance = BlobServiceClient(account_url=STORAGEACCOUNTURL, credential=STORAGEACCOUNTKEY)
+        blob_client_instance = blob_service_client_instance.get_blob_client(CONTAINERNAME, BLOBNAME, snapshot=None)
+        with open('/home/user/data/audio.mp3', "wb") as file:
+            blob_data = blob_client_instance.download_blob()
+            file.write(blob_data.readall())
         pipe.model.config.forced_decoder_ids = [[2, transcribe_token_id]]
         print("Call pipeline...")
-        text = pipe(file, return_timestamps=True)
+        text = pipe('/home/user/data/audio.mp3', return_timestamps=True)
         print("Save file...")
         file_path = '/home/user/data/new_file.txt'
         with open(file_path, "w") as file:
@@ -36,8 +48,8 @@ def transcribe(file: UploadFile):
 app = FastAPI()
 
 @app.post("/transcribe")
-def transcribe(background_tasks: BackgroundTasks, file: UploadFile = File(...)):
-    background_tasks.add_task(transcribe, file)
+def transcribe(background_tasks: BackgroundTasks):
+    background_tasks.add_task(transcribe)
     return {"text": "Processing file..."}
 
 @app.get("/text")