new inference

app.py

@@ -1,24 +1,54 @@
-from transformers import pipeline
+# from transformers import pipeline
+# import gradio as gr
+#
+# pipe = pipeline(model="dacavi/whisper-small-hi")  # change to "your-username/the-name-you-picked"
+# def transcribe(audio):
+#     text = pipe(audio)["text"]
+#     return text
+#
+# iface = gr.Interface(
+#     fn=transcribe,
+#     inputs=gr.Audio(sources="microphone", type="filepath"),
+#     outputs="text",
+#     title="Whisper Small Hindi",
+#     description="Realtime demo for Hindi speech recognition using a fine-tuned Whisper small model.",
+# )
+#
+# iface.launch()
+
 import gradio as gr
-from huggingface_hub import login
-#with open("../../token.txt", "r") as file:
-#    token = file.readline().strip()
+from transformers import pipeline
+from moviepy.editor import VideoFileClip
+import tempfile
+import os
+
+pipe = pipeline(model="dacavi/whisper-small-hi")
 
-#login(token=token, add_to_git_credential=True)
+def transcribe_video(video_url):
+    # Download video and extract audio
+    with tempfile.NamedTemporaryFile(suffix=".mp4", delete=False) as temp_video:
+        os.system(f"youtube-dl -o {temp_video.name} {video_url}")
+        video_clip = VideoFileClip(temp_video.name)
+        audio_clip = video_clip.audio
+        temp_audio_path = tempfile.NamedTemporaryFile(suffix=".wav", delete=False).name
+        audio_clip.write_audiofile(temp_audio_path, codec="wav")
 
+    # Transcribe audio
+    text = pipe(temp_audio_path)["text"]
 
-pipe = pipeline(model="dacavi/whisper-small-hi")  # change to "your-username/the-name-you-picked"
+    # Clean up temporary files
+    os.remove(temp_video.name)
+    os.remove(temp_audio_path)
 
-def transcribe(audio):
-    text = pipe(audio)["text"]
     return text
 
 iface = gr.Interface(
-    fn=transcribe,
-    inputs=gr.Audio(sources="microphone", type="filepath"),
+    fn=transcribe_video,
+    inputs="text",
     outputs="text",
-    title="Whisper Small Hindi",
-    description="Realtime demo for Hindi speech recognition using a fine-tuned Whisper small model.",
+    live=True,
+    title="Video Transcription",
+    description="Paste the URL of a video to transcribe the spoken content.",
 )
 
 iface.launch()

requirements.txt

@@ -1,3 +1,5 @@
 transformers
 torch
 tensorflow
+moviepy==1.0.3
+ffmpeg