Upload 2 files

app.py

@@ -0,0 +1,36 @@
+import streamlit as st
+from faster_whisper import WhisperModel
+import yt_dlp as youtube_dl
+
+def download_youtube_video(url: str, file_name: str) -> str:
+    ydl_opts = {
+        'format': 'bestaudio/best',
+        'outtmpl': f'{file_name}.%(ext)s',
+        'postprocessors': [{
+            'key': 'FFmpegExtractAudio',
+            'preferredcodec': 'mp3',
+            'preferredquality': '192',
+        }],
+    }
+    with youtube_dl.YoutubeDL(ydl_opts) as ydl:
+        ydl.download([url])
+        return f"{file_name}.mp3"
+
+def transcribe_audio(file_name: str):
+    model = WhisperModel("large-v2", device="cpu", compute_type="float32")
+    segments, _ = model.transcribe(file_name)
+    return ''.join([segment.text for segment in segments])
+
+st.title("YouTube Video Transcription with Whisper")
+
+url = st.text_input("Enter YouTube URL:")
+if st.button("Download and Transcribe"):
+    if url:
+        st.write("Downloading audio...")
+        audio_file = download_youtube_video(url, "podcast_audio")
+        st.write("Transcribing audio...")
+        transcription = transcribe_audio(audio_file)
+        st.write("Transcription:")
+        st.write(transcription)
+    else:
+        st.error("Please enter a valid YouTube URL.")

requirements.txt

@@ -0,0 +1,4 @@
+faster-whisper==1.0.3
+streamlit==1.24.1
+yt-dlp==2024.8.1
+pydub==0.25.1