Getting QQ Chat Data

• Tutorial reference: NTQQ Windows Data Decryption
• Supplementary material: Database Decoding Reference
• The above two are different chapters of the same tutorial, read them patiently, it's not complicated (if you don't know how, scroll to the bottom to find me)
• Use DB Browser for SQLite, enter the 16-digit key you obtained as the password
• HMAC algorithm is generally SHA1, some people use SHA512 and 256, test yourself, wrong algorithm will fail to open the database (so you need to test until it opens, you can also use AI to help you adapt)
• In DB Browser export the SQL of c2c_msg_table
• Create a new database, import the SQL file you just exported
• Get a database like this
• Structure as shown below, it's a plaintext database (you can open it and see the data, which means it's normal)
• Rename the database to qq.db and place it in the dataset/original folder

Or modify the qq_db_path in setting.jsonc

• 

Getting Telegram (TG) Chat Data

• Please use Telegram Desktop to export chat data

• Click the Export chat history button

• Select the JSON(Machine-readable JSON) button

• No need to check other buttons, as this project does not support multimodal yet

• Move all ChatExport_ folders from the export folder to the dataset/original/ folder, as shown below

• Important

• Modify the setting.jsonc file, change telegram_chat_id to your telegram chat id

Including spaces!!!

• For example, if the following ID needs to be filled in as qqqqq f

---

Getting WeChat (WX) Chat Data

• Go to the WeChatBakTool Github project and download the latest version from the releases page.

• Or click here to download WeChatBakTool.

• Go to this project to download an older version of WeChat (v3.9.12.15).

• Or click this link to quickly download WeChat.

• Install and log in to WeChat.

• On your phone, go to Settings - Chats - Chat History Migration & Backup - Migrate - Migrate to PC/Mac and proceed.

• Unzip BakTool.

• 0. Install .NET Desktop Runtime (note: this is the Desktop Runtime version 6.0; ignore if already installed).
• 1. Open and log in to WeChat.
• 2. At the bottom left of the software, click "New Workspace".
• 3. In the "New Workspace" interface, select the WeChat process for which you want to create a workspace, and confirm that the WeChat ID below is correct.
• 4. For the decryption method, it is recommended to choose "Username Inference Search"! This method theoretically supports all 64-bit versions of WeChat. However, this mode requires ensuring the WeChat account is correct.
• 5. Beginners should ignore other options and directly click "Create Workspace". The program will automatically create and decrypt the workspace.

• Right-click on the Workspace -> Manage, export friend chats, all.

• Go to the baktool folder, enter workspace-[random_folder_name]-DecDB.

• Find all MSG*.db files, for example MSG1.db, and move them all to the dataset/original/wechat folder.

(Optional) Getting Chat Data from Video/Audio Files

• Extract from dual-track video/audio (requires files with separated audio tracks)

1. Install Dependencies

# Automatically install all dependencies
python process_data/chat_parser/video-to-chatml/start-vtc.py --install

# Or install manually
pip install -r requirements.txt

Note: You also need to install ffmpeg:

• Windows: Download from https://ffmpeg.org or use choco install ffmpeg
• Ubuntu: sudo apt install ffmpeg
• macOS: brew install ffmpeg

2. Usage

Interactive Mode (Recommended for new users)

python process_data/chat_parser/video-to-chatml/start-vtc.py -i

Direct Command Line Mode

python process_data/chat_parser/video-to-chatml/start-vtc.py video.mp4 -u 0 -a 1 -o output.json -m base

Using Main Program

python process_data/chat_parser/video-to-chatml/video-to-chatml.py video.mp4 -u 0 -a 1 -o output.json -m base

Parameter Description

• video: Input video file path
• -u, --user-track: User audio track index (default: 0)
• -a, --assistant-track: Assistant audio track index (default: 1)
• -o, --output: Output ChatML file path
• -m, --model: Whisper model size (tiny/base/small/medium/large, default: base)

Whisper Model Selection

Model	Parameters	Memory Usage	Speed	Accuracy
tiny	39M	~1GB	Fastest	Lowest
base	74M	~1GB	Fast	Low
small	244M	~2GB	Medium	Medium
medium	769M	~5GB	Slow	High
large	1550M	~10GB	Slowest	Highest

Output Format

The generated ChatML file format is as follows:

[
  {
    "role": "user",
    "content": "User's spoken content",
    "timestamp": {
      "start": 0.0,
      "end": 2.5
    }
  },
  {
    "role": "assistant", 
    "content": "Assistant's response",
    "timestamp": {
      "start": 2.5,
      "end": 5.0
    }
  }
]

Common Issues

1. CUDA Support

If you have an NVIDIA GPU, the program will automatically use CUDA acceleration. Check CUDA support:

python process_data/chat_parser/video-to-chatml/start-vtc.py --check

2. Audio Track Recognition

Using interactive mode allows you to view all audio track information in the video, helping you select the correct track index.

3. Out of Memory

If you encounter memory issues, try using smaller Whisper models (like tiny or base).

Dependencies

• Python 3.7+
• openai-whisper
• torch
• ffmpeg-python
• ffmpeg (system dependency)

Supported Formats

Video Formats: MP4, MKV, AVI, MOV, WMV and other formats supported by ffmpeg Audio Codecs: Most common audio codecs (AAC, MP3, WAV, etc.)

2. [TODO] Automatic recognition and extraction from single-track video/audio (not implemented yet)