11/8/2023 0 Comments Ai channel transcripts![]() The user goes to the application's website and inputs a YouTube video link.The application displays the answer to the user through the user interface.When a user query comes in, Langchain performs a similarity search on the embedded vectors and returns the most relevant answer.The embedded vectors are stored in the Pinecone vector datastore for efficient similarity search.The text transcripts are embedded into vector format using Langchain and OpenAI embedding models.The downloaded audio is then passed to the OpenAI 'Whisper API' to convert it into text transcripts.The application uses the ytdl-core npm package to download the audio from the YouTube video.The user provides a YouTube video link through the application's user interface. ![]() Tested with a 7.5 hours long video's audio, though i did not download the audio using this code, that would have taken days to process □♀️ you can question about this video as well, the video is : Application Workflow
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |