akadeb
Re:AMD buys World Labs for $8.2B
on Tuesday September 29, 2026 @08:14AM (#49891879)
AMD prob wants a leg up when it comes to spatial models. Especially because NVIDIA is investing heavily in robotics like acq. Huggingface and Sim2Real projects
Show HN: A San Francisco cable car ride as a procedural risograph film
https://cable-car-ride.vercel.app
I saw a risograph render with Claude on Reddit and wanted to try it on the Powell/Hyde cable car. It's a 97s film covering Union sq, a view of the Bay Bridge, Coit Tower, a cafe near my old house, Lombard St and ends with a view of the Golden Gate bridge and Ghirardelli.Source, design notes and the render tools: https://github.com/akdeb/sf-cable-car-ride
Re:Show HN: Cactus Needle 3: 8-29MB automation models can match DeepSe...
on Monday September 21, 2026 @01:48PM (#49790701)
hey viccis the tool calling on ESP32 with a Mac M4 is exactly what I built here. check it out and lmk if it helps https://github.com/akdeb/open-toys
Re:Om Malik has died
on Monday June 29, 2026 @04:43AM (#48716509)
RIP legend
Show HN: Realtime Voice AI on ESP32 with Cloudflare Durable Objects
https://github.com/akdeb/ElatoAI/tree/main/server/cloudflare
Cloudflare's recent voice agents release provides a free daily tier for Deepgram STT, TTS use.So I created a Voice AI Agent pipeline that runs on Cloudflare workers with Durable Objects and it interfaces with an Arduino ESP32. This lets you bring any LLM to power your Voice AI Toys and Desk companions with STT + TTS handled natively.
Re:Show HN: I built a free CharacterAI that runs locally
on Wednesday March 25, 2026 @05:40PM (#47523652)
Yeah I sell one here https://www.elatoai.com/products/ai-devkit
If you pre-order on the link I can ship you one at a discount
Re:Show HN: I built a free CharacterAI that runs locally
on Wednesday March 25, 2026 @05:38PM (#47523627)
Thanks Boris! The case is made out of Resin but I started with PLA. Will post the open-sourced STEP files there
Show HN: I built a free CharacterAI that runs locally
https://github.com/akdeb/open-toys/tree/main/resources
CharacterAI has over 20M active users and lets people talk to AI voice clones. I built a local alternative that clones any voice with <10s of audio and can talk fluently in over 10 languages, and runs locally on your Macbook or Mac mini.The voice pipeline currently supports MLX on any M1 through M5 chip. I used Whisper-Turbo for STT, Qwen3.5-9B-4bit for the LLM and Qwen3-TTS-0.6B-4bit for...
I built a screen-free, storytelling toy with an ESP32
Re:I built a screen-free, storytelling toy with an ESP32
on Tuesday March 17, 2026 @05:31PM (#47418605)
I built an open-source, screen-free, storytelling toy for my nephew who uses a Yoto toy. My sister told me he talks to the stories sometimes and I thought it could be cool if he could actually talk to those characters in stories with AI models (STT, LLM, TTS) running locally on her Macbook and not send the conversation transcript to cloud models.
This is my voice AI stack:
- ESP32 on Arduino to interface with the Voice AI pipeline
- mlx-audio for STT (whisper) and TTS with streaming (`qwen3-tts` / `chatterbox-turbo`)
- mlx-vlm to use vision language models like Qwen3.5-9B and Mistral
- mlx-lm to use LLMs like Qwen3, Llama3.2, Gemma3
- Secure websockets to interface with a Macbook
This repo supports inference on Apple Silicon chips (M1/2/3/4/5) but I am planning to add Windows soon. Would love to hear your thoughts on the project.
Show HN: Build a $10 AI Toy that runs locally on your Mac
https://github.com/akdeb/OpenToys/tree/main/app
Hi HN! I built an open-source framework for local AI toys.With a <$10 ESP32 device and a Mac running Apple Silicon, the toy can have real-time conversations without sending any audio or text to the cloud.The project supports voice cloning, multilingual conversations (10 languages), and runs fully locally using Whisper + MLX VLMs and Qwen3-TTS and Chatterbox-Turbo.
Show HN: AI Toys that don't need the internet
https://github.com/akdeb/open-toys/blob/main/README.md
A lot of parents I speak to are concerned about sending their children's chat transcripts to the cloud and privacy is often the first thing that comes up when we talk about AI toys.So I built OpenToys so anyone with an ESP32 can create their own AI Toys that run inference locally, starting with Apple Silicon chips and keep their data from leaving their home network.The repo currently supports...
Show HN: An Open-Source Yoto Toy with Qwen3-TTS
Re:Show HN: An Open-Source Yoto Toy with Qwen3-TTS
on Saturday March 14, 2026 @04:13PM (#47380693)
I open-sourced my AI toy company last year and a lot of parents wanted their children's conversations to be private and not sent to a server.
So I created OpenToys so anyone with an ESP32 can create Yoto-like cards and stories in over 10 languages locally and create educational content with local LLMs like Mistral and Qwen3.5-9B all running on MLX.
Re:Qwen3-TTS Family Is Now Open Sourced: Voice Design, Clone, and Gene...
on Friday January 23, 2026 @05:35AM (#46730853)
You could pipe the output to an audio file with ffmpeg or pyaudio and save it locally
Re:Ask HN: Could you share your personal website here?
on Wednesday January 14, 2026 @02:39PM (#46621673)
Show HN: I Hid Labubus in World Labs' AI Worlds
https://www.akadeb.xyz/vibes/world-labubus/
I made a chrome extension game where your goal is to find a daily labubu hidden in world labs' marble worlds. Github: https://github.com/akdeb/world-labubus
Show HN: Cartoony AI Voices on ESP32 with Pitch Shifting
https://blog.elatoai.com/posts/how-to-get-ai-voices-with-different-pitch-on-esp32-arduino
I show how to use Pitch shifting supported by the arduino-audio-tools[1] and ElatoAI[2] library on ESP32 to get OpenAI Realtime and Gemini Live Voices to sound like cartoons like Alvin and the Chipmunks or Hulk.[1] https://github.com/pschatzmann/arduino-audio-tools [2] https://github.com/akdeb/ElatoAI
Re:Building Effective AI Agents
on Wednesday June 18, 2025 @07:24AM (#44308883)
I would highly recommend gemini 2.5 pro too for their speech quality. It's priced lower and the quality is top notch on their API. I made an implementation here in case you're interested https://www.github.com/akdeb/ElatoAI but its on hardware so maybe not totally relevant
Show HN: Gemini 2.5 Flash Audio on an ESP32 Using Deno
https://github.com/akdeb/ElatoAI
I ran Gemini 2.5 Flash Preview Native Audio Dialog on an ESP32 with Server VAD using Deno
Re:Human coders are still better than LLMs
on Thursday May 29, 2025 @09:53PM (#44132058)
I like the sound of that! I think youre gonna like what we are building here https://github.com/akdeb/ElatoAI
Its as if the rubber duck was actually on the desk while youre programming and if we have an MCP that can get live access to code it could give you realtime advice.
Re:I Open-Sourced My AI Toy Company That Runs on ESP32 and OpenAI Real...
on Friday April 25, 2025 @07:20AM (#43792408)
I understand, is it the realtime conversational aspect or just in general you wouldn't want a child to play with a TTS-like service?
Re:I Open-Sourced My AI Toy Company That Runs on ESP32 and OpenAI Real...
on Friday April 25, 2025 @07:18AM (#43792400)
Murphy's law
Re:I Open-Sourced My AI Toy Company That Runs on ESP32 and OpenAI Real...
on Friday April 25, 2025 @07:17AM (#43792395)
thanks for checking it out Bert
Re:I Open-Sourced My AI Toy Company That Runs on ESP32 and OpenAI Real...
on Friday April 25, 2025 @07:16AM (#43792391)
Hi Mr. teddy bear!
Hey there buddy! Have you tried brushing with Sensodyne now available at your nearest CVS only for $9.99!
Re:I Open-Sourced My AI Toy Company That Runs on ESP32 and OpenAI Real...
on Wednesday April 23, 2025 @11:44AM (#43773449)
Thank you! It's been super fun to work on. The challenges were more on the ESP32 side. Like getting audio to work smoothly with Opus and the audio timing challenges. This is one of the reasons I open-sourced.
It seems pointless to think that everyone should cross that C++/Audio barrier to make something cool. Using this cuts a lot of dev time and brings products out to market wayy quicker. The repo basically helps launch your AI toy brand
Re:I Open-Sourced My AI Toy Company That Runs on ESP32 and OpenAI Real...
on Wednesday April 23, 2025 @11:42AM (#43773417)
The emojis are all AI. The content is a mix of me n cursor and I added the mermaid chart to make it easier to visualize the system diagram.
The circuit diagram in on figma
And demo video edited on capcut
Re:I Open-Sourced My AI Toy Company That Runs on ESP32 and OpenAI Real...
on Wednesday April 23, 2025 @11:38AM (#43773374)
thank you stavros!
Re:I Open-Sourced My AI Toy Company That Runs on ESP32 and OpenAI Real...
on Wednesday April 23, 2025 @11:35AM (#43773343)
Pipecat is awesome! is it similar to what livekit provides?
I think Realtime API adoption would be higher if it is offered on Arduino rather than ESP-IDF as the latter is not very beginner friendly. That was one of the main reasons I built this repo using edge functions instead of a direct WebRTC connection.