/status

main.py

@@ -1,10 +1,12 @@
-from faster_whisper import WhisperModel
 from fastapi import FastAPI
-from video import download_convert_video_to_audio
+from fastapi.middleware.cors import CORSMiddleware
+import os
 import yt_dlp
 import uuid
-import os
-from fastapi.middleware.cors import CORSMiddleware
+from typing import Dict
+
+from faster_whisper import WhisperModel
+from video import download_convert_video_to_audio
 
 app = FastAPI()
 
@@ -16,9 +18,10 @@ app.add_middleware(
     allow_headers=["*"],
 )
 
-# or run on GPU with INT8
-# model = WhisperModel(model_size, device="cuda", compute_type="int8_float16")
-# or run on CPU with INT8
+# Create a dictionary to store the status of the current transcribing process
+# This will be updated as the process progresses
+# The key is a string describing the current process, and the value is a string describing the current status
+status: Dict[str, str] = {}
 
 def segment_to_dict(segment):
     segment = segment._asdict()
@@ -28,26 +31,28 @@ def segment_to_dict(segment):
 
 @app.post("/video")
 async def download_video(video_url: str):
+    status["video"] = "Downloading..."
     download_convert_video_to_audio(yt_dlp, video_url, f"/home/user/{uuid.uuid4().hex}")
+    status["video"] = "Downloaded"
 
 @app.post("/transcribe")
 async def transcribe_video(video_url: str, beam_size: int = 5, model_size: str = "tiny", word_timestamps: bool = True):
-    print("loading model")
+    status["model"] = "Loading..."
     model = WhisperModel(model_size, device="cpu", compute_type="int8")
-    print("getting hex")
+    status["model"] = "Loaded"
+    status["video"] = "Downloading..."
     rand_id = uuid.uuid4().hex
-    print("doing download")
     download_convert_video_to_audio(yt_dlp, video_url, f"/home/user/{rand_id}")
+    status["video"] = "Downloaded"
+    status["transcription"] = "Transcribing..."
     segments, info = model.transcribe(f"/home/user/{rand_id}.mp3", beam_size=beam_size, word_timestamps=word_timestamps)
     segments = [segment_to_dict(segment) for segment in segments]
-    total_duration = round(info.duration, 2)  # Same precision as the Whisper timestamps.
-    print(info)
+    total_duration = round(info.duration, 2)
     os.remove(f"/home/user/{rand_id}.mp3")
-    print("Detected language '%s' with probability %f" % (info.language, info.language_probability))
-    
+    status["transcription"] = "Completed"
+    status["language"] = "Detected language '%s' with probability %f" % (info.language, info.language_probability)
     return segments
 
-# print("Detected language '%s' with probability %f" % (info.language, info.language_probability))
-
-# for segment in segments:
-#     print("[%.2fs -> %.2fs] %s" % (segment.start, segment.end, segment.text))
+@app.get("/status")
+async def get_status():
+    return status