# tlack/babytalk > Fully offline speech-to-text and text-to-speech for ESP32-S3/P4 in C, MicroPython and AtomVM, with STT/TTS models and a data generation and finetuning framework. - Magnitude: 2.9 out of 10 — Early signal - Stars: 17 total · +4 stars measured 2026-10-10, 04:39–07:26 UTC, ≈ 15 by evening - Star trust: star growth looks organic - Category: Hardware, IoT & robotics · Language: C · Created: 2026-09-26 · Last push: 2026-10-09 - GitHub: https://github.com/tlack/babytalk · Page: https://gitnova.dev/en/r/tlack/babytalk ## Useful for - Transcribe speech to text on a keyboardless LoRa node and send the message - Read incoming text messages aloud on a screenless device - Build voice control that works without WiFi and never sends audio to the cloud ## Why it’s here - Star-counter measurements on 2026-10-10 (UTC), 04:39–07:26: 13 → 17 stars (+4). This is the change over that interval. - Estimated end-of-day forecast: about +15 stars, using observed gains and the previous day. - The repository is 14 days old and already has 17 stars. - Hacker News: “Show HN: Babytalk: Offline speech to text and text to speech on ESP32” — 10 points, 10 h ago. ## Star trust Star growth looks organic. Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer. ## Numbers - Forks: 3 - Issues and pull requests: 2 - Watchers: 1 - Average over the last week: 4 per day - Usual pace: 0 per day - Stars in the last hour (measured): 2 ## Stars per day, last 15 days (oldest → newest, today is partial) 2026-09-26 … 2026-10-10: 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 17, 4 ## Hacker News - Show HN: Babytalk: Offline speech to text and text to speech on ESP32 — 10 points, 5 comments: https://news.ycombinator.com/item?id=50026819 ## Spotted in now - Spotted on Hacker News ## More in this category 1. **nullmoth/nvidia-macos-driver** — 6.9 · Peaking · Hardware, IoT & robotics · Rust · +197 stars measured 2026-10-10, 00:31–07:23 UTC, ≈ 624 by evening A Metal driver for NVIDIA GeForce RTX cards on Intel Macs and OpenCore systems running macOS 15 Sequoia, letting the GPU drive the desktop, Metal, OpenCL and Core ML. Free and open source. Full card: https://gitnova.dev/en/r/nullmoth/nvidia-macos-driver.md 2. **wesbos/photo-frame-dashboard** — 6.5 · Early signal · Hardware, IoT & robotics · JavaScript · +57 stars measured 2026-10-10, 00:31–07:23 UTC, ≈ 274 by evening Notes and scripts for gaining ADB, root and firmware backups on locked-down Android photo frames (Skylight Calendar, BIUFRAME), replacing vendor software with a custom launcher and kiosk app. Full card: https://gitnova.dev/en/r/wesbos/photo-frame-dashboard.md 3. **KingKongRobotics/jumper** — 4.6 · Peaking · Hardware, IoT & robotics · Python · +30 stars measured 2026-10-10, 00:32–07:24 UTC, ≈ 101 by evening Open-source 22-DoF crab robot Jumper: this repository covers motion training, replay and policy export, plus links to tools for designing appearances and scenes. Full card: https://gitnova.dev/en/r/KingKongRobotics/jumper.md 4. **roboterax/video-prediction-policy-2** — 4.3 · Early signal · Hardware, IoT & robotics · Python · +14 stars measured 2026-10-10, 00:32–07:24 UTC, ≈ 54 by evening Official implementation of Video Prediction Policy 2, a world-action model that couples event-level video pretraining with a single-step visual planner and an action module for robot manipulation. Full card: https://gitnova.dev/en/r/roboterax/video-prediction-policy-2.md 5. **skylartaylor/gbos-vm** — 4.1 · Early signal · Hardware, IoT & robotics · Python · +6 stars measured 2026-10-10, 00:32–07:24 UTC, ≈ 31 by evening Runs a real Googlebook OS image (ARM64, Android 17) in a GPU-accelerated QEMU VM on Apple Silicon Macs, with Chrome working and Vulkan passed through to Metal. Aimed at developers who want to poke at the OS without real hardware. Full card: https://gitnova.dev/en/r/skylartaylor/gbos-vm.md --- Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score. Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast. Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself. Data as of 2026-10-10 07:34 UTC, updated every 30 minutes.