CI Build Verified · Commit 2c9befc
ESP32-S3 Conversational AI Health Voice Terminal
Voice assistant terminal combining local wake-word engine (ESP-SR), streaming ASR/TTS, DeepSeek LLM integration via SSE, dual I2S audio codecs, and LVGL graphical display.
Engineering Specifications
Microcontroller
ESP32-S3
Toolchain
ESP-IDF
Domain
Smart Home & Multimodal
CI Verification
Actions Log ↗
Hardware Deliverables & Scope
Includes hardware bench testing photos.
Clean Firmware Source Code (C/C++ PlatformIO / ESP-IDF)
Complete Pin Mappings & Hardware Schematics
GitHub Actions Continuous Integration & Build Artifacts
PCB Layout & Fabrication Gerber Files
Known Engineering Constraints & Bench Boundaries
All projects are built from real bench testing and validated by automated static analysis and continuous integration. Note the following operational boundaries:
- Local wake-word model requires ESP-SR 8 MB PSRAM octal flash configuration.
- DeepSeek SSE streaming response latency depends on outbound Internet bandwidth.
- Audio amplifier requires decoupled power filtering to eliminate DAC switching noise.