Anil Chandra Naidu Matcha publishes open-source generative-media applications and model integrations under the GitHub account Anil-matcha. His projects include an image-and-video studio, a staged short-drama generator and a real-time voice-agent example.[1][2]
Start with AI Voice Agent →
Links
GitHub Contributions
An application layer over model services
Matcha’s profile distinguishes applications in his own account from MuAPI SDKs and integrations in the SamurAIGPT organization. Open Generative AI supplies a self-hostable studio interface for generating images and video. Its open application code does not mean the underlying models run locally: the project directs model calls through MuAPI and requires that service’s credentials.[1][2]
Turning a story idea into production stages
MicroDrama AI divides creation into screenwriting, storyboarding, frame generation and video production. The repository describes passing the results between those stages and assembling clips into a short drama. This is an application pipeline connecting generative services, rather than a claim that Matcha trained the underlying media models. Its examples and requirements let developers inspect the proposed workflow before running it.[3]
A smaller voice application to inspect
AI-Voice-Agent demonstrates handling a voice call and generating a response in real time. Its setup requires Python, Deepgram and OpenAI credentials. Unlike the larger media studio, the short setup and application source give a narrower entry point for someone learning how speech and model responses fit together.[4]
Tools, writing & talks
Tools
Start here
AI Voice Agentgithub.comA self-hosted example of a voice agent using speech and language-model services.
Start with this smaller application to trace the voice-response flow. It requires Python 3.11 and Deepgram and OpenAI keys; service usage is separate from the open-source code.
Repository owner[4]Original source- Suno ComfyUImuapi.ai
Custom ComfyUI nodes for generating, remixing, and extending AI music with Suno through the muapi.ai API.
Catalog workOriginal source - AI Clipping ComfyUIgithub.com
Open-source ComfyUI pipeline that turns long videos and podcasts into ranked short clips with face-tracked auto-crop.
Catalog workOriginal source - Open Character AIgithub.com
Open-source Character.AI alternative offering preset and custom AI personas with per-chat LLM tuning.
Catalog workOriginal source - Midjourney ComfyUIgithub.com
Custom nodes that route ComfyUI graphs to Midjourney V7, V8, and Niji, with reference images and four-image runs.
Catalog workOriginal source - GPT-Image-2 ComfyUIgithub.com
Custom ComfyUI nodes for text-to-image and image-to-image generation with GPT-Image-2 via muapi.ai.
Catalog workOriginal source - Veo 3.1 ComfyUI Nodesgithub.com
Custom ComfyUI nodes for Veo 3.1 covering text-to-video, image-to-video, extend, and 4K upscale via MuAPI.
Catalog workOriginal source - AI B-Rollvadoo.tv
Tool for generating B-roll footage for videos with AI, built as a Jupyter notebook.
Catalog workOriginal source - Veo 4 APImuapi.ai
Python wrapper for Google DeepMind's Veo 4 video model with 4K output, audio, and camera controls.
Catalog workOriginal source - Open Poe AImuapi.ai
Self-hosted chat UI for GPT, Claude, Gemini, Llama and other LLMs using your own API keys.
Catalog workOriginal source - Happy Horse 1.0 API Wrappergithub.com
Python wrapper for Alibaba's Happy Horse 1.0, generating 1080p video with audio from text or images.
Catalog workOriginal source - Seedance2 ComfyUIgithub.com
Custom ComfyUI nodes driving ByteDance Seedance 2.0 text-to-video, image-to-video, and video extension locally.
Catalog workOriginal source - Open AI Micro Drama Generatorgithub.com
A staged application pipeline for scripts, storyboards, frames and short video clips.
Repository owner[3]Original source - Seedance 2.0 API Wrappermuapi.ai
Python wrapper for ByteDance's Seedance 2.0, generating realistic 1080p text-to-video and image-to-video.
Catalog workOriginal source - Open AI Design Agentmuapi.ai
Open-source autonomous design agent that generates posters, brand kits, ad creatives, and short video.
Catalog workOriginal source - Open Generative AIgithub.com
A self-hostable image-and-video studio that calls model services through MuAPI.
Repository owner[1][2]Original source