Home / Our R&D journey
Company · Since July 2025

How we built Recho: 16 months of R&D.

Recho was built in-house by Techieyan Technologies, an engineering firm in Hyderabad working on hardware, IoT and AI since 2018. This is what happened along the way.

Quick answerUpdated 27 Oct 2026

Recho has been in development since July 2025. Generic speech-to-text failed on real speech, so we built our own pipeline; accuracy rose from 71% on the first model to 95% in real-user testing in March 2026. The hardware went through two generations before the hospital device with a screen for confirming every request, launched in October 2026.

Timeline

Month by month.

Jul 2025Problem research and user interviews; product scope defined.
Aug 2025Off-the-shelf speech-to-text tested and rejected; system architecture designed.
Sep 2025Hardware v1: too large, poor audio in noise. Custom PCB needed.
Oct 2025First domain-tuned AI model at 71% accuracy; hardware v2 designed.
Nov 2025Multi-label extraction for sentences with several requests: 88% accuracy; dashboard v1.
Dec 2025First end-to-end system; 5–8 second delay found and targeted.
Jan–Feb 2026Pipeline optimised, 95% accuracy, delay cut sharply.
Mar 2026Real-user testing: 95 of 100 spoken inputs understood correctly.
Apr 2026Hardware finalised; DPIIT Startup India recognition.
May–Sep 2026Focus moved to hospitals; hospital workflow built; demo to a major hospital group.
Oct 2026Hospital device with screen-confirm, Record and Send keys; rechotechnologies.com relaunched.
Lessons

What we learned.

Generic speech AI is not enough. Names, places and casual speech broke off-the-shelf models, so we built a domain-tuned pipeline.

People say several things at once. One sentence often holds two or three requests, so Recho splits them into separate tasks.

Trust needs a screen. Showing the words before sending, with a re-record option, is what makes staff comfortable relying on it.

Hardware is worth it. A dedicated device beats a phone app for habit: no unlocking, no notifications, one button to speak.

FAQ

Questions people ask.

Who builds Recho?
Recho is built by Recho Technologies, part of Techieyan Technologies, an engineering firm in Hyderabad that has delivered hardware, IoT and AI projects since 2018.
How accurate is Recho’s speech recognition?
In real-user testing in March 2026, 95 of 100 spoken inputs were understood and categorised correctly. Every request is also shown on screen to the sender before it is sent.
Why did Recho move from health tracking to hospitals?
The same technology, a press-and-speak device with an AI that structures casual speech, solves a bigger and clearer problem in hospitals, with a clear buyer.
Is the hardware designed in-house?
Yes. Electronics, PCB layout, firmware, software and most enclosure design are done in-house; board fabrication and some prototyping are outsourced.
Related

See what 16 months built.

A 15-minute demo of the device, the routing and the live board.