Speech & Audio
Transcription, TTS, voice agents. The key principle is privacy: processing happens locally, a minimum leaves the machine.
What we do
Transcription
Whisper running locally: voice messages, meetings, interviews — into text with speaker diarization.
Speech synthesis
TTS agents: voiceovers, voice assistants, multilingual support.
Voice agents
The full loop: heard → understood → answered by voice. Training, support, language practice.
Voice interfaces
Voice as a product channel: support assistants, orders and notifications by voice, interactive scenarios for phone lines and call centers.
Products
RecordAI
Meeting recording on a Mac in two channels → local whisper.cpp transcription on Metal → live translation and summaries via your n8n. Audio never leaves the computer.
Voice English Tutor
A voice AI English tutor: live spoken dialogue, instant error breakdown, practice history. Built on edge-tts + Whisper + LLM.
Have a project in this area?
Tell us what you need — we will suggest a format: subscription, turnkey deployment or a lab pilot.