PRAGUE (AFP) - Vlastimil Gular’s existence took an unwelcome turn a 12 months in the past: minor surgery on his vocal cords found out throat cancer, which led to the loss of his larynx and with it, his voice.

however the fifty one-yr-antique father of 4 remains chatting away using his very own voice in preference to the tinny timbre of a robot, way to an progressive app advanced by means of two Czech universities.

"I discover this very beneficial," Gular advised AFP, the use of the app to kind in what he wanted to mention, in his personal voice, via a mobile smartphone.


"I’m now not excellent at the usage of the voice prosthesis," he introduced, pointing at the hole the scale of a big coin in his throat.

This small silicon device implanted within the throat lets in humans to talk by means of pressing the hole with their arms to alter airflow through the prosthesis and so create sound.

however Gular prefers the brand new hi-tech voice app.

It become developed for sufferers set to lose their voice due to a laryngectomy, or elimination of the larynx, an ordinary process for superior ranges of throat cancer.

The joint mission of the university of West Bohemia in Pilsen, Prague’s Charles university and two personal companies -- CertiCon and SpeechTech -- kicked off nearly two years in the past.

The era makes use of recordings of a affected person’s voice to create synthetic speech that can be played on their cellular telephones, capsules or laptops via the app.

ideally, sufferers need to report more than 10,000 sentences to offer scientists with enough cloth to produce their artificial voice.

"We edit collectively individual sounds of speech so we want a variety of sentences," stated Jindrich Matousek, an professional on text-to-speech synthesis, speech modelling and acoustics who heads the project on the Pilsen college.

‘A be counted of weeks’ 

but there are drawbacks: sufferers going through laryngectomies typically have little time or strength to do the recordings inside the wake of a analysis that calls for quick remedy.

"It’s typically a depend of weeks," stated Barbora Repova, a health practitioner on the Motol college health center, operating on the undertaking for Charles college.

"The patients also ought to tackle issues like their economic situation, their lives are grew to become the other way up, and the ultimate element they need to do is to make the recording," she instructed AFP.

To deal with those problems, scientists came up with a more streamlined technique for the app, which is supported by means of the era enterprise of the Czech Republic.

operating with fewer sentences -- preferably 3,500 however as few as 300 -- this technique makes use of advanced statistical fashions together with synthetic neural networks.

"you operate speech fashions with positive parameters to generate synthesised speech," stated Matousek.

"Having more statistics continues to be better, however you could acquire respectable pleasant with much less facts of a given voice."

The sentences are cautiously selected and individual sounds ought to be recorded several times as they're suggested in a different way next to one-of-a-kind sounds or at the beginning and quit of a word or sentence, he delivered.

so far, the Pilsen university has recorded 10 to fifteen sufferers, according to Matousek.

besides Czech, the Pilsen scientists have additionally created synthesised speech samples in English, Russian and Slovak.

‘baby dinosaurs’ 

Gular -- an upholsterer who lost his activity due to his handicap -- controlled to document 477 sentences over the 3 weeks among his diagnosis and the operation.

but he become confused and less than happy with the exceptional of his voice.

"Throat most cancers patients regularly be afflicted by some form of dysphonia (hoarseness) before the surgical procedure, so in mixture with a constrained speech pattern it makes the voice sound unnatural," said Repova.

In a studio on the Pilsen college in the meantime, entrepreneur Jana Huttova is recording outlandish phrases.

The 34-yr-antique mother of 3 faces the danger of dropping her voice to minor throat surgical treatment -- an operation on her parathyroid gland.

"The Chechens have usually desired a dagger-like Kalashnikov," she says, reading from the text earlier than her.

"i've small kids and i want them to pay attention my personal voice, now not a robot," Huttova stated.

Then she moved directly to her subsequent sentence: "We have been attacked by a tyrannosaur’s infant dinosaurs."

related to the mind 

Matousek believes that within the destiny, patients could be capable of use the app to report their voice at home the usage of a specialised website to manual them via the manner.

And he hopes that at some point it's going to move even in addition.

"The ultimate vision is a miniature tool related to the mind, to the nerves linked to speech -- then patients could manage the device with their thoughts," he said.

This sort of superior answer is a totally long way off, stated Repova.

"but have a look at cochlear implants -- forty years ago once they commenced, we had no idea how it would increase, how extensively they could turn out to be being used," she stated, relating to the internal-ear implants used to address intense deafness.

"A happy end would be a tool implanted within the throat that might speak with the patient’s very own voice," she advised AFP.

"It’s sensible: it can now not are available a 12 months or maybe in 10 years, however it’s realistic and we’re on the way."