Update app.py

app.py

@@ -1,7 +1,7 @@
+import spaces
 import torch
 
 import gradio as gr
-import spaces
 import yt_dlp as youtube_dl
 from transformers import pipeline
 from transformers.pipelines.audio_utils import ffmpeg_read
@@ -9,7 +9,7 @@ from transformers.pipelines.audio_utils import ffmpeg_read
 import tempfile
 import os
 
-MODEL_NAME = "openai/whisper-large-v3"
+MODEL_NAME = "ylacombe/whisper-large-v3-turbo"
 BATCH_SIZE = 8
 FILE_LIMIT_MB = 1000
 YT_LENGTH_LIMIT_S = 3600  # limit to 1 hour YouTube files
@@ -23,6 +23,7 @@ pipe = pipeline(
     device=device,
 )
 
+
 @spaces.GPU
 def transcribe(inputs, task):
     if inputs is None:
@@ -71,7 +72,7 @@ def download_yt_audio(yt_url, filename):
         except youtube_dl.utils.ExtractorError as err:
             raise gr.Error(str(err))
 
-
+@spaces.GPU
 def yt_transcribe(yt_url, task, max_filesize=75.0):
     html_embed_str = _return_yt_html_embed(yt_url)
 
@@ -94,14 +95,13 @@ demo = gr.Blocks()
 mf_transcribe = gr.Interface(
     fn=transcribe,
     inputs=[
-        gr.Audio(type="filepath"),
+        gr.Audio(sources="microphone", type="filepath"),
         gr.Radio(["transcribe", "translate"], label="Task", value="transcribe"),
     ],
     outputs="text",
-    theme="huggingface",
-    title="Whisper Large V3: Transcribe Audio",
+    title="Whisper Large V3 Turbo: Transcribe Audio",
     description=(
-        "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the OpenAI Whisper"
+        "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the"
         f" checkpoint [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to transcribe audio files"
         " of arbitrary length."
     ),
@@ -111,14 +111,13 @@ mf_transcribe = gr.Interface(
 file_transcribe = gr.Interface(
     fn=transcribe,
     inputs=[
-        gr.Audio(type="filepath", label="Audio file"),
+        gr.Audio(sources="upload", type="filepath", label="Audio file"),
         gr.Radio(["transcribe", "translate"], label="Task", value="transcribe"),
     ],
     outputs="text",
-    theme="huggingface",
     title="Whisper Large V3: Transcribe Audio",
     description=(
-        "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the OpenAI Whisper"
+        "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the"
         f" checkpoint [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to transcribe audio files"
         " of arbitrary length."
     ),
@@ -132,10 +131,9 @@ yt_transcribe = gr.Interface(
         gr.Radio(["transcribe", "translate"], label="Task", value="transcribe")
     ],
     outputs=["html", "text"],
-    theme="huggingface",
     title="Whisper Large V3: Transcribe YouTube",
     description=(
-        "Transcribe long-form YouTube videos with the click of a button! Demo uses the OpenAI Whisper checkpoint"
+        "Transcribe long-form YouTube videos with the click of a button! Demo uses the checkpoint"
         f" [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to transcribe video files of"
         " arbitrary length."
     ),