The person behind it
Louis-Paul Baril founded and sold technology businesses, and led AI automation projects inside Québec organizations. As a producer at HELPY Media, Louis-Paul runs live events and teaching: church services from the audio console, conferences, corporate events, courses, and live streams followed around the world. Live Caption Studio comes out of that work.
Where it started
The idea grew one event at a time, seen from the production booth.
People who could not hear the room
It started in the audience. At services and conferences, people who are deaf or hard of hearing held up a phone and tried to follow the speaker with a transcription app, its microphone catching the room from across the hall. Meanwhile the sound console had a clean feed of every voice. The room should be the one giving them the words, on its screens and on their own phones.
Four languages, one stage
Then came the multilingual events. For a Siemens event, our team at HELPY Media delivered the program in four languages (French, English, Portuguese and Spanish) with interpreters, for the people in the room and for viewers following the live stream around the world. Getting every language to every audience at the same moment took a lot of coordination.
Audio that should not leave the room
Private corporate events and internal training raised another question: where does the audio go? A product launch before its announcement, or a course built on a company’s own know-how, should not travel to a cloud service to be transcribed. So Live Caption Studio does the work on a computer in the room.
After the event
Version 1.0 is planned to keep the transcript of every session and, when you turn it on, to record the voices on that same computer. A later 1.x release plans an after-event module that works only from that local record: a clean transcript of the whole event, meeting minutes with a summary, decisions and action items, short highlights to promote your next event, social media post drafts, and a searchable archive. Nothing is sent to the cloud, and every output is a draft you review before anything is published.
Why we build it to run locally
Live Caption Studio is built to transcribe and translate on your own computer, and it never sends audio to a cloud service. The only network calls are license activation or renewal and update checks, and both can be turned off. We decided this early, because it is the first thing the rooms we build for ask about.
Our design partner
We build with a church in Québec where Louis-Paul runs the audio console. Its sound team routes the feeds we design against, and it will be the first room to run the beta on real services. We will name it when it gives permission.
How we test
We publish measured results before we claim them. Delay and accuracy will be published with the date, the hardware and the method, and we call a language supported only after our evaluation passes for it. Until then, every page says “planned”, and we quote no number we have not measured.
Company facts
- Legal name
- Neural Twiin
- Address
- 5450 rue Jarry Est, bureau 100, Montréal (Québec) H1P 1T9
- Contact
- [email protected]
A short form: your room, your gear, your languages. No payment.