DEDUPEDWho it is forA content creator, teacher, or course author who has a long Vietnamese script in chapters or parts and needs one audio file per part on a personal computer.
The problemRecording or generating a long script in one go is slow, and a single failure forces a full redo. Splitting by hand is tedious.
The outcomeEach section of the script becomes its own WAV or MP3 file, all downloadable individually or as one .zip, with no internet connection needed after install.
What you need first
- A Windows 10/11 64-bit, Ubuntu/Debian 64-bit, or macOS 13/14 computer, with at least 8 GB RAM and 3 GB free disk space (5 GB recommended)
- The text to read, as a TXT, PDF, or Word (.docx) file, or pasted directly into the app
- For the subtitle step, an .srt or .vtt subtitle file for the video
Steps
- Download the installer for your system from the Releases page at https://github.com/Cuongyd196/cit-voice-studio/releases/latest (or the Google Drive link in the README if GitHub is slow).
- Install and open CIT Voice Studio. On Linux, run
sudo apt install ./CIT-Voice-Studio-v1.0.3-amd64.deband then open CIT Voice Studio from the app menu. - Open the Tạo hàng loạt tab, then paste the script or upload a TXT file. Put a marker such as [P1] and [P2] at the start of each line that should become its own file.
- Choose a voice and speed, then start generation. Keep the app open until it finishes.
- Download each file one by one, or download all of them in one .zip file.
Prompt for your agent
Using the CIT Voice Studio app already installed on this computer, take the script in <script-file-path>, insert [P1], [P2] and so on at the start of each section, and run it through the Tạo hàng loạt tab. Expected output: one WAV or MP3 file per section plus a .zip of all files. Verify that the number of output files equals the number of [P] markers in the script.
Caveat. The app keeps working only while it is open, and the README says the Linux and macOS builds show a Terminal window or browser tab that must stay open during use.
Use this in your AI agent
Ask your agent
Using the Deduped MCP, get the setup steps for Cuongyd196/cit-voice-studio and "Generate separate narration audio files from one text file"
Add Deduped to Claude Code once
claude mcp add --transport http deduped https://dedupedai.com/mcp
Every Monday: the 5 hottest repos and exactly how to use them. Free, one email a week, unsubscribe anytime.
Get them freeShare: X LinkedIn Hacker News Email Copy link
Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer