MAKING FROM ZERO
I am Qihao. I started without a manufacturing background and am learning, with help from AI, how to make physical things. I document the trials, failures, and improvements that turn each idea into something real.
CURRENT CHAPTER
The site grows with the real work. This section always keeps the latest public experiment, problem, or milestone.
How one ESP32-S3 prototype moved from a complete but sluggish record–transcribe–respond–synthesize pipeline to one real-time voice session—and why the hard part was coordinating audio, networking, display, and playback on constrained hardware.
Read this updateTHE PROCESS OUTLASTS THE ANSWER
Power, structure, materials, and even how a problem is described can stop an idea. Writing the problem clearly keeps the next attempt from repeating it.
How one ESP32-S3 prototype moved from a complete but sluggish record–transcribe–respond–synthesize pipeline to one real-time voice session—and why the hard part was coordinating audio, networking, display, and playback on constrained hardware.
A postmortem of severe ESP32-S3 recording undersampling: from a Base64 buffer error to identifying low-speed ST7789 redraws that blocked I²S, then validating the fix with a dedicated FreeRTOS capture task and rate-limited partial refreshes.
Documents the tested ESP32-S3 and INMP441 pipeline, from press-and-hold recording, on-screen volume feedback, and WAV upload to server-side audio processing, whisper.cpp transcription, and web display.
BUILDS THAT LASTED
These ideas survived testing and rework and became things I continue to use.
A completed, button-triggered AI voice conversation prototype: the ESP32-S3 streams microphone audio into one real-time session and returns AI replies through its screen and speaker.
START AT THE BEGINNING
This archive follows each build from the first attempt through failure, rework, and a working result. Start with the earliest public note.
