model selection added..

.gitignore

@@ -128,4 +128,6 @@ dmypy.json
 # Pyre type checker
 .pyre/
 *.mp3
-*.mp4
+*.mp4
+
+.DS_Store

app.py

@@ -20,8 +20,8 @@ def inference(audio):
     print(result.text)
     return result.text
 
-def inference2(url):
-    return main(url)
+def inference2(url,mymodel):
+    return main(url,mymodel)
 
 
 title="YouWhisper"
@@ -89,6 +89,23 @@ css = """
             font-weight: bold;
             font-size: 115%;
         }
+        .gr-dropdown {
+            background-color: #fff;
+            border: 1px solid #e5e5e5;
+            border-radius: 4px;
+            color: #000;
+            cursor: pointer;
+            display: inline-block;
+            font-size: 1rem;
+            font-weight: 400;
+            line-height: 1.5;
+            margin-bottom: 0;
+            padding: .375rem .75rem;
+            transition: border-color .15s ease-in-out,box-shadow .15s ease-in-out;
+            vertical-align: middle;
+            white-space: nowrap;
+            width: 100%;
+        }
 """
 
 block = gr.Blocks(css=css)
@@ -158,8 +175,11 @@ with block:
 
         # add another textbox to get url from user
         url = gr.Textbox(label="URL", show_label=False, placeholder="Enter Youtube URL")
+        # add dropdown to select the model
+        mymodel = gr.Dropdown(["tiny", "base", "small", "medium"])
         btn2 = gr.Button("Transcribe")
         text2 = gr.Textbox(show_label=False, placeholder="Transcription will appear here")
-        btn2.click(inference2, inputs=[url], outputs=[text2])
+        btn2.click(inference2, inputs=[url, mymodel], outputs=[text2])
+
         
 block.launch()

convert.py

@@ -21,11 +21,13 @@ def AudiotoText(filename):
     sonuc = result["text"]
     return sonuc
 
-def main(url):
+def main(url,mymodel):
     print('''
     This tool will convert Youtube videos to mp3 files and then transcribe them to text using Whisper.
     ''')
 
+    print("URL: " + url)
+    print("Model: " + mymodel)
     print("Downloading video... Please wait.")
     try:
         filename = download_video(url)
@@ -35,12 +37,12 @@ def main(url):
         return
     try:
         convert_to_mp3(filename)
-        print("Converted video to mp3")
+        print("Video converted to mp3")
     except:
         print("Error converting video to mp3")
         return
     try:
-        model = pywhisper.load_model("base")
+        model = pywhisper.load_model(mymodel)
         result = model.transcribe(filename[:-4] + ".mp3")
         print(result["text"])
         result = result["text"]
@@ -49,10 +51,9 @@ def main(url):
         print("Removed video and audio files")
         print("Done!")
         return result
-    except:
-        print("Error transcribing audio to text")
+    except Exception as e:
+        print(e)
         return
     
-
 if __name__ == "__main__":
     main()

requirements.txt

@@ -3,5 +3,4 @@ whisper
 pytube
 moviepy
 pywhisper
-flask
-static-ffmpeg
+flask