Subtitles for the Real World

An Indian and an American student come together to co-found a start-up that creates affordable assistive technology for those hard of hearing.

By Jason Chiang

October 2023

Subtitles for the Real World

TranscribeGlass is a wearable assistive device for the deaf, hard of hearing, the elderly and people who want to better understand spoken communication. (Photograph courtesy TinkerTech Labs)

By 2050, nearly 2.5 billion people are projected to have some degree of hearing loss, according to the World Health Organization. There are about 63 million people in India suffering from significant auditory impairment, says an estimate by WHO. The impact from a growing communication barrier for those who are hard-of-hearing and the deaf population can be significant—like isolation from society, higher unemployment or lower wages.

Madhav Lavakare, a Yale University student from New Delhi and Tom Pritsky, a Stanford University graduate who suffers from hearing loss in both ears, co-founded TranscribeGlass, a start-up which has created an affordable solution for people with auditory impairments. The product, which too is called TranscribeGlass, is a wearable assistive device for the deaf, hard of hearing, the elderly and people who want to better understand spoken communication. The versatile smart glass device translates speech-to-text in real time and overlays it onto a small transparent display in the user’s direct field of vision. Unlike speech-to-text applications on devices, where users have to look down at a screen to follow the conversation, TranscribeGlass allows users to read captions while looking at the speaker.

To use TranscribeGlass, the user launches its mobile app and connects to a caption source. For example, Google API or Deepgram. The text is transmitted via Bluetooth to the TranscribeGlass display. Aiming for wide usage, it was designed to weigh less than 30 grams and can be attached to a pair of glasses or even an empty frame. Users can also change the caption font size and TranscribeGlass currently supports any language that uses Roman/Latin alphabets.

Joining forces

Lavakare was in grade 11 in New Delhi when a friend, who was hard of hearing, dropped out of his school because he was not able to follow lessons or conversations around him. Lavakare wanted to find a solution to help his friend. “Mobile apps can transcribe speech-to-text in real time but since my friend heavily depended on visual cues, there was a need to bring captions into his direct field of vision,” says Lavakare, who received a grant from the U.S.-India Science & Technology Endowment Fund (USISTEF) in 2020.

Pritsky, founder of Stanford University’s club for the deaf and hard of hearing, has had hearing loss since the age of 3 and regularly uses hearing aids and lip-reading to communicate. Like Lavakare, Pritsky too envisioned the possibility of an affordable captioning device to help himself and others with hearing loss. “I really like captions for movies,” says Pritsky. “I thought it would be fantastic to have them for real life.”

Both Lavakare and Pritsky, who met through a mutual friend, have close ties to the world of assistive tech and were independently pursuing the idea before joining forces. According to Lavakare, running the business on his own was hard as a full-time student at Yale, making Pritsky a much-needed addition to the team. Lavakare adds that Pritsky’s insights as a user was a “powerful perspective” to have at the start-up.

The prototype

Feedback from users has been a crucial part of their design process. Lavakare and Pritsky estimate that over the course of its various prototype iterations, TranscribeGlass has had at least 300 people test their product.

After completing five different prototypes, TranscribeGlass is now ready to offer an affordable and comfortable alternative to existing solutions. “One of the things that we really value with TranscribeGlass is the social acceptability of the solution,” says Lavakare. “Our goal is to be source agnostic,” adds Pritsky. “We can integrate any API: Google Speech, Deepgram, Microsoft. Our larger goal is to help solve hearing loss.”

The company recently began manufacturing its first 150 preorders, with hopes to finish shipping them in the next few months. The TranscribeGlass Beta is being sold for about Rs. 4,500, with the final version expected to be priced around Rs. 8,000.

“I also see there being a lot of upgrades in future versions of TranscribeGlass,” says Lavakare, “including having the ability to translate languages or potentially having graphics that could help people with other disabilities.”.

Jason Chiang is a freelance writer based in Silver Lake, Los Angeles.

Click here to sign up for the free SPAN newsletter:


Leave a Reply

Your email address will not be published. Required fields are marked *