community: add video transcript loaders using Whisper for enhanced video transcription

[TeodorZlatanov]

Description:
This pull request introduces six new loader classes to the community package, enhancing video processing capabilities across different environments using the Whisper model. These classes facilitate the transcription of video files into either segmented or paragraphed text formats. Each loader class returns comprehensive information about the transcription, including the source, start time, end time, and other relevant metadata. The classes are designed to cater to different setups: Azure, Local, and OpenAI platforms, each offering both paragraph and segment-based processing.

New Classes:

• AzureWhisperVideoParagraphLoader: Processes video files into paragraphs using Azure's Whisper API.
• AzureWhisperVideoSegmentLoader: Processes video files into segments using Azure's Whisper API.
• LocalWhisperVideoParagraphLoader: Transcribes local video files into paragraphs using the local Whisper model.
• LocalWhisperVideoSegmentLoader: Transcribes local video files into segments using the local Whisper model.
• OpenAIWhisperVideoParagraphLoader: Utilizes OpenAI's cloud-based Whisper API to transcribe videos into paragraphs.
• OpenAIWhisperVideoSegmentLoader: Utilizes OpenAI's cloud-based Whisper API to transcribe videos into segments.

Dependencies:

• Whisper: Required for local loaders. Install via pip:

  pip install openai-whisper
  

• OpenAI: Required for Azure OpenAI and OpenAI loaders. Install via pip:

  pip install openai
  

• FFmpeg: Required for preprocessing video files into audio formats that Whisper can process. Install FFmpeg:

  # on Ubuntu or Debian
  sudo apt update && sudo apt install ffmpeg
  
  # on Arch Linux
  sudo pacman -S ffmpeg
  
  # on MacOS using Homebrew (https://brew.sh/)
  brew install ffmpeg
  
  # on Windows using direct download:
  Download from https://ffmpeg.org/download.html and add the executable to your PATH.
  
  # on Windows using Chocolatey (https://chocolatey.org/)
  choco install ffmpeg
  
  # on Windows using Scoop (https://scoop.sh/)
  scoop install ffmpeg

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Comments	Updated (UTC)
langchain	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	May 15, 2024 7:20am

[TeodorZlatanov]

@hwchase17 , just wanted to bring this PR to your attention.