basic gradio app, init!

app.py

@@ -0,0 +1,47 @@
+import gradio as gr
+import whisper
+from pytube import YouTube
+
+loaded_model = whisper.load_model("base")
+
+def inference(link):
+  yt = YouTube(link)
+  path = yt.streams.filter(only_audio=True)[0].download(filename="audio.mp4")
+  options = whisper.DecodingOptions(without_timestamps=True)
+  results = loaded_model.transcribe(path)
+  return results['text']
+
+def change_model(size):
+  loaded_model = whisper.load_model(size)
+
+title="Youtube Whisperer"
+description="Speech to text transcription of Youtube videos using OpenAI's Whisper"
+block = gr.Blocks()
+
+with block:
+    gr.HTML(
+        """
+            <div style="text-align: center; max-width: 500px; margin: 0 auto;">
+              <div>
+                <h1>Youtube Whisperer</h1>
+              </div>
+              <p style="margin-bottom: 10px; font-size: 94%">
+                Speech to text transcription of Youtube videos using OpenAI's Whisper
+              </p>
+            </div>
+        """
+    )
+    with gr.Group():
+        with gr.Box():
+          sz = gr.Dropdown(label="Model Size", choices=['base','small', 'medium', 'large'], value='base')
+          sz.change(change_model, inputs=[sz], outputs=[])
+          link = gr.Textbox(label="YouTube Link")
+          text = gr.Textbox(
+              label="Transcription", 
+              placeholder="Transcription Output",
+              lines=5)
+          with gr.Row().style(mobile_collapse=False, equal_height=True): 
+              btn = gr.Button("Transcribe")       
+              btn.click(inference, inputs=[link], outputs=[text])
+
+block.launch()

requirements.txt

@@ -0,0 +1,3 @@
+transformers
+git+https://github.com/openai/whisper.git
+pytube