Update tools/transcription_tool.py

tools/transcription_tool.py

@@ -1,25 +1,43 @@
-from youtube_transcript_api import YouTubeTranscriptApi
+import yt_dlp
 import re
-from smolagents.tools import Tool
-class YoutubeTranscriptions(Tool):
-    name = "YoutubeTranscriptions"
-    description = "Generates transcription for a YouTube video and summarize them. if transcription cannot be generated, just say no transcriptions available. Do not call any other tools"
-    inputs = {'URL': {'type': 'string', 'description': 'URL of the video to transcribe'}}
-    output_type = "string"
 
+class YoutubeTranscriptions:
+    def __init__(self):
+        self.name = "YoutubeTranscriptions"
+        self.description = "Generates transcription for a YouTube video. If transcription cannot be generated, return 'No transcriptions available.'"
+    
     def forward(self, URL: str) -> str:
         # Extract video ID from URL
         video_id_match = re.search(r"v=([a-zA-Z0-9_-]+)", URL)
         if not video_id_match:
             return "Invalid YouTube URL. Please provide a valid URL."
 
-        video_id = video_id_match.group(1)
+        video_url = URL
+
+        # yt-dlp options to get subtitles
+        ydl_opts = {
+            'quiet': True,
+            'skip_download': True,
+            'writesubtitles': True,
+            'subtitleslangs': ['en'],  # Adjust if you want other languages
+            'format': 'bestaudio/best'
+        }
 
         try:
-            transcription = YouTubeTranscriptApi.get_transcript(video_id)
-            if transcription:
-                return transcription
+            with yt_dlp.YoutubeDL(ydl_opts) as ydl:
+                info = ydl.extract_info(video_url, download=False)
+                subtitles = info.get('subtitles', {})
+
+                if 'en' in subtitles:
+                    subtitle_url = subtitles['en'][0]['url']
+
+                    # Download and extract subtitles
+                    import requests
+                    response = requests.get(subtitle_url)
+                    if response.status_code == 200:
+                        return response.text  # Returns the actual transcription
+
+                return "No transcriptions available."
         except Exception as e:
             return f"Error fetching transcription: {str(e)}"
 
-        return "Sorry, the transcription was not available."