In his later years, Stephen Hawking’s voice became as famous as his ground-breaking theories on quantum gravity, black holes, and theoretical cosmology. Considered the most famous theoretical physicist of his generation, Hawking greatly advanced our understanding of black holes, predicting their ability to emit thermal radiation (now known as Hawking radiation). However, it was Hawking’s image and voice that caught the public imagination as a true symbol of the triumph of mind over matter. It is well known that Hawking was diagnosed with Motor Neurone disease – also known as amyotrophic lateral sclerosis, a disease of the nerve cells in the spinal cord and brain that are responsible for voluntary movement – in his twenties. Following an emergency tracheotomy in 1985, Hawking completely lost the ability to speak, which he described in his biography:
For a time, [after the tracheotomy] the only way I could communicate was to spell out words letter by letter, by raising my eyebrows when someone pointed to the right letter on a spelling card. It is pretty difficult to carry on a conversation like that, let alone write a scientific paper.
This was when the work of American scientist Dennis Klatt came in. A speech synthesis pioneer, Klatt made several advances in the generation of high-quality human speech by machine. For example, his formant synthesis software, published in the Journal of Acoustical Society of America in 1980, is still used today by scientists concerned with the study of synthesising speech sounds. Klatt also contributed to the development of the text-to-text speech system known as the “KlattTalk” TTS (text-to-speech) system. This was widely used by those with disabilities, particularly people with low vision who used this system to read aloud computer text. One of these voices was Klatt himself, known as “Perfect Paul”, and became famous as the voice of Professor Stephen Hawking. In fact, Hawking became so attached to Klatt’s voice that he considered it a part of his identity – upon being offered newer versions, such as the Speech Plus synthesiser in 1988, he preferred to continue with Klatt’s. Thus, technology gave a voice to Hawking, who would have otherwise suffered from locked-in syndrome, a disorder in which a person is trapped within their body without being able to express their self. Hawking praised the apparatus:
This system allowed me to communicate much better than I could before …Using this system, I have written a book, and dozens of scientific papers. I have also given many scientific and popular talks …I think that this is in a large part due to the quality of the speech synthesiser.
Klatt had been working on speech synthesis since the 1960s at the Massachusetts Institute of Technology (MIT), where he developed his algorithm, KlattTalk. Klatt’s colleague and research scientist during the 1970s and 1980s, Joseph Perkell, described his work as exceptionally good, and praised his hard-working ethic: ‘He can function in the middle of a hurricane.’ Klatt spent hours generating human-like speech from text fed into his computer, using his own voice for the sound of “Perfect Paul”. ‘For him, the most convenient sound was the sound of his own voice,’ Perkell says. He also created a female version known as “Beautiful Betty”, and a child version, “Kit the Kid”, based on his daughter, Laura. In 2013, Laura shared her experiences with speech synthesis, stating that Klatt was most concerned with making sure his system was natural and intelligible. At the time, Klatt’s system was certainly of the highest quality, however it was still described by The New York Times as “like a scratchy recording of a person with a lisp.” Even in the present, there is still much to be done to improve the naturalness of computer-generated speech.
Other technicians and scientists used KlattTalk to create a mobile computer for Professor Hawking’s wheelchair. Notably, KlattTalk was developed commercially by Digital Equipment Corporation in 1983 and became known as DECtalk. Hawking’s speech was first controlled with a hand switch that allowed him to select words by moving a cursor through options on his computer screen. Later modifications used infrared beam switches which responded to his head and eye movements only. Until 2014, Hawking used the DECtalk speech synthesiser, before switching to a newer system which outputted a more realistic-sounding voice, the Speech Plus CallText 5010 synthesiser. Despite the change in system, he still used Perfect Paul’s voice: “I keep it because I have not heard a voice I like better and because I have identified with it.” His voice became so recognisable that Hawking even gave permission for Perfect Paul to be used in his 2014 biopic, The Theory of Everything. Recordings of his voice have also appeared in many songs, such as “Keep Talking” by Pink Floyd and “Fitter Happier” by Radiohead.
Hawking’s voice became his trademark, however Perkell always associated it with Klatt. He told Witness History: ‘I thought, wow, I’m watching Stephen Hawking and out comes Dennis’ voice.’ Unfortunately, Klatt was diagnosed with thyroid cancer in the early 1980s and the treatment gradually caused his laryngeal tissues to stop functioning. ‘It’s ironic. He focused so much on reproducing his natural voice that he lost it in himself’, Perkell notes. Dr Klatt continued to work despite his illness and remained optimistic until his death at age 50 in 1988. His voice, however, lived on. Perkell remarked that ‘It was never the case that when I heard Hawking speaking that I didn’t think of Dennis.’ Thus, Klatt’s legacy lived on through Hawking in a way that certainly reflects Klatt’s proudest achievement in speech synthesis.
Stephen Hawking’s ability to communicate via synthetic speech certainly contributed to his longevity as a productive scientist, and this is largely thanks to Dennis Klatt. Hawking’s story is not inspiring only because of his impressive theories in physics, but because of his ability to persevere despite his difficult circumstances. His voice demonstrates how severe speech impediment can be overcome through computer technology and illustrates the obstacles and issues that people with speech impairments experience. There are still many issues regarding the technology of speech synthesis that must be addressed in the future of Computerised Speech Synthesis (CSS), such as the circumstances surrounding a speech impairment, and the intelligibility of CSS voices, however the developments surrounding speech synthesis for assistive aid remain promising. Undoubtedly, the most famous example of speech synthesis, Stephen Hawking, must be remembered alongside the man who gave him his voice, Dennis Klatt.
Written by Kat Jivkova
Bibliography:
Carr, Bernard J, et al. “Stephen William Hawking CH CBE. 8 January 1942-14 March 2018.” The Royal Society Publishing. [Accessed on 23 February 2022].
Handley, Rachel. “Stephen Hawking’s voice, made by a man who lost his own.” [Online]. [Accessed on 23 February 2022].
Kewley-Port, Diane, and Terrance M. Nearey. “Speech Synthesizer Produced Voices for Disabled, Including Stephen Hawking.” The Journal of the Acoustical Society of America 148, no. 1 (2020): R1–R2.
Mullennix, John, and Steven Stern. “From Wood to Bits to Silicon Chips: A History of Developments in Computer Synthesized Speech.” In Computer Synthesized Speech Technologies: Tools for Aiding Impairment, 9–27. IGI Global, 2010.
Pinner, Richard S. “History of Augmented Communication: Technology and Disability.” In Augmented Communication, 21–27. Cham: Springer International Publishing, 2018.
Retrospect Journal 