AB
AiBoss
project

BILIVE - an open-source AI recording tool for one-stop automatic live streaming recording and processing.

BILIVE is an open-source tool based on AI technology, specifically designed for recording and processing Bilibili live streams. The tool supports automatic recording of live streams, rendering of bullet comments and subtitles, speech recognition, automatic segmentation of exciting clips, and generating interesting...

What is BILIVE?

BILIVE is an open-source tool based on AI technology, specifically designed for recording and processing live streams on Bilibili. The tool supports automatic recording of live streams, rendering of bullet comments and subtitles, speech recognition, automatic segmentation of highlights, and generation of interesting titles and stylized video covers. BILIVE can automatically upload processed videos to Bilibili. It integrates multiple modal models, is compatible with ultra-low-configuration machines, and can run without a GPU, making it suitable for personal users and small servers.

BILIVE's main functions

  • Live recordingAutomatically records Bilibili live streams, supports simultaneous recording in multiple rooms, and includes bullet comments and gift information.
  • Bullet comments and subtitlesAutomatically renders bullet comments and supports voice recognition to generate subtitles.
  • Automatic slicingExtract highlights based on the density of bullet comments and generate titles.
  • Cover generationAutomatically generates stylized video covers.
  • Automatic submissionSupports uploading videos to Bilibili, and is compatible with multi-part uploads.
  • Low hardware requirementsNo GPU required, suitable for running on very low-spec machines.

How to use BILIVE

  • Cloning projectRun the following commands in the terminal to clone the project and initialize submodules:
git clone --recurse-submodules https://github.com/timerring/bilive.git
cd bilive
  • Install dependenciesInstall Python dependencies in the project directory, ensuring that ffmpeg is installed:
pip install
-r requirements.txt
  • Configuration and Startup:
    • Core configurationSet the API Key and mode for Automatic Speech Recognition (ASR), Multi-Layer Model (MLLM), and Image Generator. Configure video upload information (title, description template, section tid, etc.).
    • Recording configurationAdd the room number (room_id) to be recorded. Set parameters such as recording duration (duration_limit).
    • Log in:
      • Log in using bilitool:
bilitool login --export
      • Or log in in a submodule:
cd src/upload/bilitool
python -m bilitool.cli login
    • Set a password for the recording page.Set the environment variable RECORD_KEY.
export
RECORD_KEY=your_password
    • Start recording:
./record.sh
    • Start processing and uploading:
./upload.sh

BILIVE's project address

Application scenarios of BILIVE

  • UP master content creationAutomatically record live streams, generate bullet comments, subtitles, clips, and covers, and quickly upload them to Bilibili.
  • Multi-platform streamingSupports 24/7 continuous streaming and simultaneous management of multiple live streaming rooms.
  • Extracting highlightsIntelligent identification of high-energy segments, generating attractive titles to help promote content.
  • Low configuration friendlyNo GPU required, suitable for older devices, and automatically cleans up to save space.
  • Study and ResearchThe recorded data can be used for analysis, and the open-source code facilitates technical learning.