Researchers from two Czech universities developed an app for throat cancer patients who are set to lose their voice as a result of a laryngectomy, the surgical removal of the larynx. Patients can still use their voice while chatting rather than the tinny timbre of a robot through the newly-developed app.
Researchers from the University of West Bohemia in Pilsen and Charles University in Prague, as well as two private companies - SpeechTech and CertiCon - started the joint project about two years ago. Patients can record their voice to make a synthetic speech, and it can be played on their laptops, tablets, or smartphones through the app.
According to the South China Morning Post, patients should record over 10,000 sentences to allow researchers to gather enough material to make their synthetic voice. Researchers need a lot of sentences because they edit together individual sounds of speech, said Jindrich Matousek, who heads the project at the Pilsen university.
However, the app has some drawbacks. Patients undergoing laryngectomies have no time and energy to record their voice in the wake of a diagnosis that requires swift treatment. To address this issue, researchers came up with a more streamlined method for the app - backed by the Technology Agency of the Czech Republic.
The method uses advanced statistical models like the artificial neural networks; it ideally needs 3,500 but as few as 300 sentences. Researchers said the sentences were selected carefully and individual sounds were recorded several times as they were pronounced differently next to different sounds or at the beginning and end of a word or sentence.
Matousek said Pilsen university had already recorded 10 to 15 patients, while Pilsen scientists also made synthesized speech samples in Slovak, English, and Russian.
Vlastimil Gular, who had minor surgery on his vocal cords which revealed throat cancer, is using the app. He said the app was very helpful and he can type in what he wants to say - in his voice - through his mobile phone. Gular had already recorded 477 sentences in three weeks between his diagnosis and the operation.
In the future, Matousek believes patients can use the app to record their voice at home using a specialized website to guide them through the process. He hopes that everything will be possible one day, their ultimate vision is to develop a miniature device that's connected to the brain, to the nerves linked to speech. Through this, patients can control the device with their thoughts.
