LingvoHelmet universal cybertranslator - connects people, breaks language barriers, embodies the ideas of science fiction! Now every descendant of Adam and Eve will understand another without obstacle! Do you want to call to Rio de Janeiro, but don't know Portuguese? Do you go on vacation to an exotic country? Do you plan to spend secret talks without involving an interpreter? Then certainly your way leads to us!

Dear visitors of "Indiegogo", hello!

We invite you to consider the following fact. There are so many different people, opinions, events... Modern communications - Internet, telephone, fax, television, radio, etc. allow to talk to each other to such different people from such different countries and localities that are only a hundred years ago and to think about it could not - Zulu and Icelander, for example, or Japanese and Portuguese.

But here's the thing - all these people are still speaking hundreds and hundreds of different languages​, which , according to legend , were mixed by Heaven during Tower of Babel construction, in order to prevent people to raise it up to the sky. And while people who want to chat , do not always know other languages ​apart from their mother tongue to speak at a sufficient level. People at different times have found to solve this problem – hiring special dragomans , creating mixed languages - ​”pidgins” and international “lingua franca”, but all this did not solve the problem . Nowadays, the market began to appear different translation tools , but their common drawback is attachment to a computer - desktop or portable.

The essence of the project.

Our team has decided to offer to mass users a "LingvoHelmet" portable device for recognition, synchronous translation and synthesizing of human speech, which will make much easier communication between two people speaking different languages. While the sides are able to communicate among themselves as live (for example, a tourist trip), as in a network, by telephone, and so on.

This instrument provides two modes of operation - "helmet2man" and "helmet2helmet". At the first mode a man wearing a device, is talking with a man who have not a device. LingvoHelmet detects the language in which an interlocutor talks with unit's owner, and translates his speech to LingvoHelmet's owner synthesized by his mother tongue (this setting LingvoHelmet stores by default and if it's necessary changes according the situation). The owner answers , the helmet simultaneously translates the response and voices it to party without a helmet by "machine" voice.

In the second mode, when both interlocutors are wearing helmets, the speech synthesis is turned off, the device translates only a helmet companion's speech to its owner. Good soundproofing of headphones provides better translated speech audibility for the device's user in comparison with the original sound, just as is the case of films dubbing , due to a higher volume level.

Extraneous noises and sounds will be filtered with both software and hardware (on the principle of sound frequency and amplitude distinguishing), so that the device will not fail the owner even in a noisy city street.

Additional information and options.

The device will be able to connect to the computer and smartphone through Bluetooth, Wi-Fi and USB. In the premium configuration LingvoHelmet will be equipped with advanced features - a video camera, AM/FM radio, a slot for a SIMM-cards, LCD – Screen, GPS, infrared port... The owner of such device would be envied by the inspector Gadget!

The device will have two speakers and two spaced microphones - to capture owner's speech and interlocutor's speech. It also will be equipped by a rendered speaker for dubbing of synthesized speech addressed to an interlocutor of the device owner. The brain of the device will be a mini-computer with ARM-processor running the Linux operating system and using free software.

Charge to be missed for 5-6 hours of continous talk. (Planned capacity of the battery 4000-5000 milliamp-hours) Сharger will be provided. (Available for wireless charger).

Dozens of languages will be available to the user ​​- from English to Swahili.

The device will adjust the speech defects.

The device will be compact and lightweight, reliable, easy.

Software that translates speech to text or action will be available (applicable in robotics). Version of the device for people with disabilities will be provided.

The device will be w/affordable price, different colors and designs to choose from (possible individual order).

Become an indispensable assistant for the translator, traveler, salesman, any person who communicates with the speakers of other languages​​.

Can be a good gift for birthday, Christmas, and so on.

With our device also will be communicate by telephone or Skype, using the speakerphone or applying habitual way of telephone or mobile phone to the right of the head outside of the device, while the microphone of your telephone will fall on LingvoHelmet's speaker and earpiece will match with LingvoHelmet's microphone.

Over time we are going to complement the speech recognition system by recognition of lipreading, using a built-in camera with the help of MakGurk effect(interaction between hearing and vision in speech perception).

Also will be available in premium device set recognition and translation of text visually reading from paper or monitor using embedded video camera.

What's our Status?

We currently have a fully functioning prototype of LingvoHelmet that works as shown. This was completely developed by our team. (Festival/Julius synthesizing-recognition + KSokrat/Mova translation + Raspberry based motherboard + Raspbian/Pidora OS + ARM 11 + RAM 512M + 4 languages: English, Spanish, French, German). W/3D printed ABS or PLA plastic body.

All your contributions will go to the first batch of products. Your support will enable us to finalise the device's new motherboard development and the acoustic model for human speech recognition-synthesizing.

The device specification.

• ARM Cortex-A8 or A9(is available in premial set) Processor;
• Linux based operating system;
• 512MB or 1 GB(is available in premial set) RAM;
• 32GB internal storage;
• 802.11 b/n/g Wi-fi Enabled (WEP, WPA, and WPA2 encryption);
• 4 Hi fidelity spaced speakers and 2 spaced microphones;
• 2,5" LCD display(is available in premial set);
• 8 or 11 megapixel camera;
• USB 2.0 with 5 V power supply;
• Bluetooth 4.0;
• Ethernet port;
• Infrared port;
• SD card slot;
• Li-ion Battery, 4500 mA/h

The process of devices development.

Production timeline.

Risks and challenges.

There may be some problems with the recognition of human speech when the speaker excited, patter says, tongue-tied, with foreign accent on some of the local dialects and jargons, in a figurative sense, and so on.

Problems may occur during a conversation with several people talking at the same time - we plan to share this kind of speech on several streams in timbre and volume of speaking voices and to translate each stream separately by different machine voices and if necessary to cut off the excess voices with a weaker sound.

May be difficult to transfer the intonations of human speech - questioning, exclamatory, and others. Synthesized voice may appear somewhat cybernetic and metal as it happens in science fiction films, although it may seem funny to many people.

FAQ

A: - Yes, by http protocol using a graphical interface, as well as you do with a home router or network printer. In the premium version a small LCD touch screen will be available.

Q: - Do not be too slow working and overheating ARM processor? Will not hang-ups and pauses in speech translation?

A: - No way! Power of ARM-9 and ARM-11 would be enough to handle the audio stream, its recognition, translation and speech synthesis. In the premium version of the device carrying additional features, we also plan to use more powerful processors Cortex. And for CPU cooling, we made a special air duct for ventilation.

Q: - How long will it to be missed batteries?

A: - 5-6 hours of continuous talk time without recharging.

Q: - Will there be different colors and designs of the device are available?

A: - All imaginable colors and styles from cyberpunk and steampunk to classical and baroque, from military and alien to road and air ... For complete sets of premium will be possible personal order.

Q: - Will be enough solid the device's body?

A: - LingvoHelmet's body will be made from impact-resistant, elastic and viscous plastic (polycarbonate), durable and resistant to breaks and scratches. For premium models will be available other materials - polished aluminum, wood and others.

Q: - Can I use it as an accessory to smartphone either as regular headphones?

A: - Yes, via USB, audio port, infrared port or bluetooth.

Q: - Will LingvoHelmet provide search function of device if it is lost or stolen?

A: - Yes, for the premium version with connection to GPS and mobile telephony.

Q: - Not bad that I wear this thing on my head? Not whether it irradiates the head and the whole body? (Just do not take me for one of the "foil hats" :-)

A: - No more harmful than conventional headphones. In addition, the Wi-Fi and Bluetooth can be disabled by using the touch LCD.

Q: - What is the warranty period for the product?

A: - Year warranty, as well as on any computer and peripherals to it, with no signs of opening and mechanical damage. For premium version a lifetime warranty is provided.

Q: - Will be the device water-resistant and dustproof?

A: - Dustproof - yes. Water-resistant – just in a special “Naval” version.

Q: - What will be the weight of the device?

A: - 200-300 grams depending on the configuration.

Q: - Whether the sex of the speaker in translation?

A: - Male voice will translate the male speech and female voice – the female speech, with recognition by the frequency characteristics of the voice. Over time, we also think to make correction for age of the speaker.

Q: - Are the translation applications for smartphones complete substitute your decision?

A: - Absolutely not, because our unit in order to best capture and playback of speech uses 4 (!) spaced speakers and 2 microphones and applies improved acoustics, which, however, did not in any smartphone; in addition do not have to explain once again that the smartphone memory constantly loaded weight of other tasks, processes and threads, and live speech translation using such applications will always be "slow down" and "sag" and not be synchronous.