speechmatics

Fast text based video editing, node Electron Os X desktop app, with Backbone front end.

Python SDKs for Speechmatics APIs

🏆 Voice AI Hack London 2026, Overall Winner. Patients minimise. Voices don't. Real-time clinical voice intelligence that flags when what a patient says diverges from what their voice reveals. Built on Speechmatics medical STT, Thymia voice biomarkers, and Claude.

Commander is a conversational background AI agent that uses Speechmatics for real time voice transcriptions and Gemini APIs for desktop automation and agentic workflow orchestration.

speechmatics test

Adversarial multi-agent audit for police bodycam footage. Prosecution, Defense and Judge agents run on different model families; every verdict is traceable to an utterance index and a video timestamp. Four jurisdictions. Prize winner - AI Agent Olympics, Milan AI Week 2026.

Real-time speech transcription & translation for live events — listeners follow on their own phones in their own language. 9 STT / 12 translation / 6 TTS pluggable engines, fully offline (whisper.cpp + NLLB) or cloud (Speechmatics, Google, DeepL…). Windows desktop + cross-platform Docker edition (EveryTongue Lite).

Voice-powered developer research assistant. Dictate a question, get a cited answer from the web in real time. React, TypeScript, Vite, Tailwind, Supabase (Auth, Postgres, Realtime, Edge Functions), Speechmatics STT, Bright Data.

A dual-lab AI project: Vault-AI – banking MCP server with LangGraph ReAct agent + Streamlit UI; Voice-AI-Bot – multimodal chatbot using Google Gemini and Speechmatics STT/TTS. Shared Python environment.

Adversarial multi-agent investment due-diligence. Milan AI Week / lablab.ai hackathon. SEC EDGAR + earnings call audio + Bull/Bear/Reconciler agents on Gemini 2.5 Pro + Featherless Qwen3 + Speechmatics, hosted on Vultr.

Home Assistant Speechmatics speech-to-text integration

Frontend part for VOCALI test project

Self-hosted video-to-text and audio-to-text transcription with speaker diarization, timestamps, batch processing, translation, AI summaries, and subtitle exports.

AI-powered voice-first disaster response platform for earthquakes, floods, cyclones & wildfires. Victims record distress calls; AI triages urgency & dispatches rescue teams in real-time.

Web app experimenting with Speechmatics real-time transcription service.

ATLAS — Enterprise multi-agent system with governance. 29/29 tests PASS. 5 sponsors integrated (Speechmatics/Featherless/Gemini/Vultr/Kraken). SOUF AI DPI + Ed25519 audit chain. AI Agent Olympics 2026.

Voice-to-Ops — field reports that write themselves. Prep repo (verified prototype + build playbook) for the AI Factory native.builder hackathon.

Frontend part for VOCALI test project

WebSocket Speechmatics Flow bridge for AVR: receives PCM audio from avr-core and streams STS audio responses in real time.

Public case study: VivaReady — AI-powered Irish-language oral exam coach. Node.js POC → Swift/iOS app → React landing page (vivaready.com).