Spectacles for the deaf using Arduino

Spectacles for the deaf using Arduino
0

Aim: The basic aim is to create a kind of SMART Spectacles for the deaf that will convert speech to text and display it for them to read.

Currently I am working on a project to make spectacles for the deaf. The following are the components I will be using.

  1. Arduino
  2. A sound detection module sensor
  3. SIM800L GSM/GPRS Module
  4. OLED Display for Arduino

The plan for my project currently is the following:

  1. Record sound and take the output in Analog.
  2. Send that output to a free website that I will create to convert it into text using the GSM/GPRS module which will provide access to the internet.
  3. Send the text back to Arduino.
  4. Display the text on the OLED Display for the Arduino which will be mounted on the spectacles.

In addition to above to take full advantage of the GSM/GPRS Module I am thinking about implementing calling and messaging facility too in the spectacles.

I would like to know what the community here at STEMGames think about it and what changes do they suggest to make this a success. I am kind of newbie to a project of this scale and scope so some of my points above might seem crazy so I welcome any suggestions and points you have to say on this.

2 Likes

Too slow to stream speech. You will require a 4G or 3G module.
http://kaldi-asr.org/doc/dependencies.html

1 Like

I would research myself on using a 4g module but do you have any suggestions as in which one should I buy? How much increase in budget for the project I must expect?

It’s a great idea.

However, what will the weight of the digital hardware be, the processor module and the OLED display? Locating this on the spectacle arm will be a critical consideration, as heavy spectacles quickly become very tiring to wear.

Second, instead of the OLED display, perhaps a projector that will display the text in reflection off the spectacle might work better, and avoid both the weight and the cantilever needed for a display, and also be much more discreet in appearance. If the projectors are light enough, the text could be cast on both spectacles, and deliver a more readable 3D print.

Third, as an alternative, or an addition, a vibrator on the spectacle arm that might directly be detected by the skullbone would help a deaf person who is also visually impaired. This could offer both Morse and TapToTalk as haptic codes.

For such persons, there is a different possibility that can be implemented, a mouth pad, to be held on the tongue. This could provide haptic sensations in Braille. Work on such pads was conducted by NCBS about 15 years back, and materials have improved a lot since then. Braille is presently preferred in India, as it is part of the curriculum, where Morse is not.

1 Like

What you are saying can be done but I will need a lot of help. I have myself put together a prototype. It doesn’t has all the features that you mentioned and it isn’t like the best version. It’s kind of bulky, poorly finished, poor design. It’s working none the less. instead of going for a costly module I have gone for direct use of mobile by pairing the specs with Bluetooth. I created a app which takes the speech input from any form of input device connected with the mobile including it’s inbuilt mic. It converts speech to text using google’s speech recognizer then it sends the text to the specs and it is displayed on an OLED display which is adjustable so that we can view it properly according to one’s convenience. I will soon give updates on it. I will soon post more detailed closeup photos of it.

Capture
Capture2

It just won me first prize in an intercollegiate science exhibition. I am thinking of improving this idea. I had pondered previously on the idea of a small projector. But I was enable to find one. What would be it’s cost and which one should I buy?

2 Likes

I don’t know whether one can get a projector module that pairs with a phone, using Bluetooth. But it certainly seems like a great way to go, since, with the phone, you already have a readymade processor and programmable platform.

1 Like

I haven’t been able to find a projector module after a very through search on google. If you too haven’t heard about it, then I think that something like that is non-existent currently. What do you suggest that I do to create a small projector that could fit inside a glass? How should I proceed researching about making one?

I have found that what I am looking for is actually an extremely small pico projector. I think it’s way out of my league to be making. One can’t buy one so that’s out of the question. @vvcstemplay What do you suggest should I go about learning about something so advanced? What sources should I refer?

My guess is that some things in micro electronics are difficult to fabricate, as they require a lot of precision. Yes, I had heard of pico projectors, but not about whether the circuit includes a Bluetooth module.

A diy alternative could be a slightly larger, and hopefully much cheaper, display projector located in the arm near the ear, with the arm itself used as an optical guide — a fiber optic cable with a lens — this might also be a simple ball of glass or plastic — glued at one end. The projection is onto the spectacle glass, so no screen is needed.

The prototype need not actually be mounted on a spectacle frame until all the optics and mechanics are closer to final.

1 Like

The mobile is already present and has a display. The additional display is merely introducing complications.
Headsup display that are not highly obtrusive is extremely hard to get right and used in super tiny niche areas.

3 Likes

Could you explain the reason?

The gadget used by optometrists to measure the eye optics has a self adjusting focus. I think one should read up on how it works. The same logic ought to make it possible to create a ready focused text line projected on the spectacle glass.

1 Like

Optometrist machine uses a pulsed laser to measure the distance and a motorised lens. The desktop machines costs several lakhs. Dont know of a non-desktop one, although it would be possible.

1 Like

Attention diversion. Hands free functionality is required while negotiating public places, where one requires to pay close attention visually - roads, stairs, objects in path etc - while the headsup display diverts ones attention, as it always is in our field of view (but at a different focal point).

My suggestion would be to make a phone application making the whole project relatively straight forward. The display being an addition for a later day.

1 Like

Ya but if we can include addition of 4G sim card network in the device it will be a fully usable speech operated phone inbuilt with specs without a need of any external device. According to me I would definitely like to have a specs with inbuilt functionalities of a phone. That is cheaper than the phone (Cost estimate around 3k max). Even smart watches are a thing so why not this? Why integrate a phone with specs just to see notifications. If you can build a phone in a pair of spectacles. I think this can be marketed as better smart glasses. And ultimately making things is fun so why not make one.

What about a normal oled display like one in the above pictures of the specs but one which can be retracted back and taken out whenever one needs.

You cannot build even half a phone at the price of full phone, because VOLUMES. The volumes you will make will probably reach 1/1000th of an alpha prototype of a commercial phone.

Commercial considerations not withstanding, your cost would be approx 10K in volumes of 100 for a phone whose specs match that of a Rs.3000/- phone.

The 2 statements are incompatible.
Making for fun as well as functionality is wonderful.

Marketing? That is a wholly different game.

As a prototype yes. As a gadget usable by hearing impaired many ifs and buts.

Let’s meet tomorrow at HBCSE and see what would be the best direction(s) to pursue.

1 Like

Even though all of your math is right the point you are missing is that when I talk about phone I am talking about calling and messaging the essence of a phone and not a android phone. Even though android systems can be run using raspberry pi if I am not wrong. There are some small versions of raspberry pi like zero w that could be used in specs.

What time would it be better for you?

I know marketing is different but why can’t one keep in mind the future yet however remotely possible largely improbable :sweat_smile: prospects of ones work in mind while also having fun doing it?

I should reach by 1130

1 Like