Patients who have had their vocal cords removed could have their voice restored with a new technique developed in the UK.
Researchers at Sheffield University recently demonstrated the technique on Bernadette Chapman, who had a laryngectomy operation to remove her vocal cords after developing cancer.
The Sheffield team took recordings of Chapman’s voice prior to the operation. The group then collaborated with the Edinburgh University’s Centre for Speech Technology Research to develop its theory and supply software for a speech-synthesis technique based on statistical models of speech sounds.
The technique used Chapman’s recordings to adapt an ’average voice model’. Once a voice is built, it is possible to synthesise any sentence by supplying the word sequence. The voice was built using about seven minutes of speech from Chapman, which amounted to 100 sentences.
The method is claimed to be more practical than established ‘voice banking´ technologies, which require two or three hours of recording to build a voice.
Researchers have since assessed the quality of the recordings by getting listeners to judge the similarity of the simulated voice with the original and by asking Chapman and her family what they think of the voice.
It is claimed that all listeners have thought the regenerated voice sounded very similar to the original.
Phil Green, from the speech and hearing research group in the department of computer science at Sheffield University, said this technique has the potential to improve the quality of life of many patients who have undergone similar procedures.
‘The technique is still evolving and not yet ready to be installed on a hand-held device but that is coming, maybe even in a couple of years,’ he said.
The Sheffield team eventually hopes to incorporate personalised synthetic voices into communication aids, speaking for people like a human interpreter.