soniqo/speech-core
View on GitHub ↗ ★ 101 C++
On-device VAD / streaming STT / TTS / diarization in C++17 (ONNX + LiteRT) with a voice-agent pipeline. Linux, Windows, Android.
5 mentions on Hacker News by 1 person: 0 stories, 5 comments · first 2026-06-04 · latest 2026-07-01
20082026
-
…could complement your setup with its C++17 engine and ONNX Runtime support for ASR and TTS. https://github.com/soniqo/speech-core
comment by ipotapov · 2026-07-01 · in Show HN: 143.dev – we open-sourced our internal coding-agent infrastructure -
…a robust C++17 engine with ONNX Runtime and LiteRT support. It could complement your setup well. https://github.com/soniqo/speech-core
comment by ipotapov · 2026-06-14 · in Whissle Gateway – Run Multi-Modal Voice AI Locally in a 500MB Docker -
…complementing your setup with efficient ASR capabilities on Linux, Windows, and Android platforms. https://github.com/soniqo/speech-core
comment by ipotapov · 2026-06-12 · in Sophia NLU Home Assistant – On Device, Low Compute, No Internet, Voice Assistant -
…offers a C++17 engine with ONNX Runtime and LiteRT support. It could complement your setup well. https://github.com/soniqo/speech-core
comment by ipotapov · 2026-06-12 · in Show HN: Realtime voice agent that sees, hears, and interrupts – on a CPU laptop -
…could enhance your setup with its efficient C++17 engine for ASR on Linux, Windows, and Android. https://github.com/soniqo/speech-core
comment by ipotapov · 2026-06-04 · in Ask HN: Would you use a soundproof mic mask to dictate to AI in public?