using gunicorn

app.py

@@ -1,7 +1,15 @@
+from asyncio.log import logger
+import logging
 import os
 import whisper
 from flask import Flask, jsonify, request
 from werkzeug.utils import secure_filename
+try:
+    import gunicorn.app.base
+    gunicorn = True
+except ImportError:
+    logging.exception("gunicorn not installed")
+    gunicorn = False
 
 STARTING_SIZE = 'small'
 UPLOAD_FOLDER = 'uploads'
@@ -9,6 +17,7 @@ ALLOWED_EXTENSIONS = {'ogg', 'mp3', 'mp4', 'wav',
                       'flac', 'm4a', 'aac', 'wma', 'webm', 'opus'}
 normal_size = STARTING_SIZE
 small_size = 'base'
+PORT = 7860
 
 app = Flask(__name__)
 app.config['UPLOAD_FOLDER'] = UPLOAD_FOLDER
@@ -18,6 +27,22 @@ model_en = whisper.load_model(f"{normal_size}.en")
 model_small = whisper.load_model(small_size)
 model_small_en = whisper.load_model(f"{small_size}.en")
 
+if gunicorn:
+    class StandaloneApplication(gunicorn.app.base.BaseApplication):
+        def __init__(self, app, options=None):
+            self.options = options or {}
+            self.application = app
+            super().__init__()
+
+        def load_config(self):
+            config = {key: value for key, value in self.options.items()
+                      if key in self.cfg.settings and value is not None}
+            for key, value in config.items():
+                self.cfg.set(key.lower(), value)
+
+        def load(self):
+            return self.application
+
 
 def inference(audio_file, model=model, model_en=model_en):
     audio = whisper.load_audio(audio_file)
@@ -84,4 +109,14 @@ def small():
 
 
 if __name__ == "__main__":
-    app.run(host="0.0.0.0", port=7860)
+
+    options = {
+        'bind': f'0.0.0.0:{PORT}',
+        # 'workers': 4,
+        'timeout': 600,
+    }
+    try:
+        StandaloneApplication(app, options).run()
+    except:
+        logging.exception("Error starting server, backing off to debug mode")
+        app.run(host="0.0.0.0", port=PORT)