Cassette
Cassette is designed to create 30-second explanatory videos suitable for Instagram Reels or YouTube Shorts.
Install / Use
/learn @M3rcuryLake/CassetteREADME
Cassette
Cassette is a Python program designed to create 30-second explanatory videos suitable for Instagram Reels or YouTube Shorts without ever leaving the terminal. It offers multiple customization options for creating personalized videos. The program utilizes APIs and libraries, including GPT-3.5-turbo model for transcript generation and UnrealSpeech API for voiceover generation, and ffmpeg along with moviepy for video editing. The 'seewav' module in the given codebase is a modified version of a pull request by @Phoenix616 at the github page of the base seewav module
Also, you may call this a indirect free python interpretation of Brainrot.js (Also an inspiration)

Video Example :
https://github.com/M3rcuryLake/Cassette/assets/105872630/e7751f0a-085e-4898-8b3c-69007c551b2f
Before Execution
Before running Cassette, follow these steps:
-
Sign Up at UnrealSpeech: Visit UnrealSpeech to create accounts.
-
Clone the Repo: Clone The repo with
git clone https://github.com/M3rcuryLake/Cassete.git -
Replace the variable
UnrealSpeech_APIinapi_keys.jsonwith your respective API key. -
Install Prerequsites: Ensure that Python and pip are installed on your system. Install additional dependencies and the Fonts by running:
sudo apt-get install -y python3-dev libasound2-dev ffmpeg pip install -r requirements.txt mkdir ~/.local/share/fonts cp fonts/* ~/.local/share/fonts/ && fc-cache -f -vif you are using windows, make sure python-pip and winget is already installed and set to path, then open any terminal (git-bash, powershell or cmd) and type in the following commands :
winget install ffmpeg pip install -r requirements.txtThen install the fonts in the /fonts/ directory manually.
Once these steps are completed, you can execute Cassette with python3 main.py or python main.py to generate your customized 30-second explanatory videos. Enjoy creating engaging content with Cassette!
Customisation Options
-
Background Music Options: Cassette allows you to add background music to your videos.
-
Choose a Voice: Select a voice for the voiceover generation.
-
Choose a Background Gameplay: Decide on the background gameplay for your video.
-
Choose a Character Image: Lastly, choose a character image for your video.
-
Subtitle styles : It allows customised timestamps (word or sentance)
-
Choose Custom Fonts : Select from multiple fonts and colours for your subtiltes.
-
Choose your own Background Colours : Allows you to choose from multiple options for background colours
Common problems/errors :
- dependencies are not resolved
- unrealspeech free tier API limit exeeded (250000 chars / 6 hrs cap on API)
- unrealspeech not responding to API Calls (common)
- g4f not responding to API calls (common)
- Correct options not chosen
- Fonts not installed
Tested Only on Linux (Ubuntu 22.04, Fedora 40)
Related Skills
TrendRadar
49.4k⭐AI-driven public opinion & trend monitor with multi-platform aggregation, RSS, and smart alerts.🎯 告别信息过载,你的 AI 舆情监控助手与热点筛选工具!聚合多平台热点 + RSS 订阅,支持关键词精准筛选。AI 智能筛选新闻 + AI 翻译 + AI 分析简报直推手机,也支持接入 MCP 架构,赋能 AI 自然语言对话分析、情感洞察与趋势预测等。支持 Docker ,数据本地/云端自持。集成微信/飞书/钉钉/Telegram/邮件/ntfy/bark/slack 等渠道智能推送。
mcp-for-beginners
15.5kThis open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable, and secure AI workflows from session setup to service orchestration.
Anthropic-Cybersecurity-Skills
3.5k734+ structured cybersecurity skills for AI agents · MITRE ATT&CK mapped · agentskills.io open standard · Works with Claude Code, GitHub Copilot, OpenAI Codex CLI, Cursor, Gemini CLI & 20+ platforms · Penetration testing, DFIR, threat intel, cloud security & more · Apache 2.0
Auto-claude-code-research-in-sleep
2.7kARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.
