AI Restores Voice to Musician Battling ALS, Returning Him to the Stage
A London performance marked a poignant return to music for Patrick Darling, who lost his ability to sing two years ago to amyotrophic lateral sclerosis (ALS). Thanks to groundbreaking artificial intelligence, Darling is composing and performing again, offering a beacon of hope for those facing similar challenges.
The Devastating Impact of Motor Neuron Disease
At 32, Patrick Darling’s life took an unforeseen turn with a diagnosis of amyotrophic lateral sclerosis (ALS), a progressive neurodegenerative disease also known as motor neuron disease (MND). ALS affects the nerves controlling muscle movement, gradually eroding the ability to walk, speak, and even breathe. Darling first noticed subtle changes a few years prior, initially dismissed as clumsiness, but these quickly escalated into a debilitating loss of physical function.
“He just kept slipping and falling, tripping on paving slabs and things like that,” recalls bandmate Nick Cocking, recounting a rainy walk in Cardiff that foreshadowed the challenges to come. By August 2023, Darling required a seat during performances, and within months, he lost the use of his hands, rendering his beloved guitar and banjo unplayable. His final performance with the Ceili House Band in June 2024 marked a heartbreaking end to a cherished chapter.
The emotional toll was immense. “He called me the day after and said he couldn’t do it anymore,” Cocking shared, his voice thick with emotion. Darling, a musician since the age of 14, proficient in bass guitar, acoustic guitar, piano, melodica, mandolin, and tenor banjo, had always found his greatest joy in singing. His partnership with Cocking, beginning over a decade ago during his university years, had elevated the musicianship of the Ceili House Band, as Cocking fondly remembers. The Ceili House Band remains a testament to their shared passion.
But Darling refused to surrender his musical spirit. He explored “voice banking,” a process of recording one’s voice to create synthetic speech, but found the technology inadequate as his voice had already changed due to the progression of the disease.
The Promise of AI-Powered Voice Cloning
Hope arrived in the form of ElevenLabs, an AI company pioneering voice cloning technology. Richard Cave, a speech and language therapist at University College London and a consultant for ElevenLabs, introduced Darling to a tool capable of recreating realistic voices from even minimal audio samples. ElevenLabs’ impact program provides free licenses to individuals who have lost their voices due to conditions like ALS, head and neck cancer, or stroke.
“We’re not really improving how quickly they’re able to communicate, or all of the difficulties that individuals with MND are going through physically, with eating and breathing,” explains Gabi Leibowitz, a speech therapist leading the program. “But what we are doing is giving them a way … to create again, to thrive.” This technology allows individuals to maintain their professional lives and, crucially, to continue expressing themselves.
Cave worked closely with Darling, utilizing existing recordings – even those captured on mobile phones in noisy pubs – to reconstruct his voice. The resulting clone wasn’t perfect, retaining a slight raspiness present in the original recordings, but it was undeniably him. “The first time I heard the voice, I thought it was amazing,” Darling shared, using his AI-generated voice. “It sounded exactly like I had before, and you literally wouldn’t be able to tell the difference.” He playfully admitted the first word his new voice uttered began with ‘f’ and ended in ‘k.’
ElevenLabs has further expanded its capabilities with Eleven Music, an AI music generator. This tool allows users to compose tracks using text prompts, selecting musical styles with ease. The company has even partnered with artists like Michael Caine to clone their voices for projects, and recently released an album of AI-generated tracks, including a lauded rendition of a Liza Minnelli song. The Liza Minnelli track is a testament to the tool’s potential.
Darling and Cave spent six weeks refining a new song, utilizing Eleven Music’s text-prompt capabilities. The result was a deeply personal composition, a heartfelt tribute to his great-grandfather, whom he never had the chance to meet.
What does this breakthrough mean for the future of artistic expression for those facing debilitating illnesses? And how might AI continue to bridge the gap between physical limitations and creative potential?
A Triumphant Return to the Stage
The culmination of this journey arrived at the ElevenLabs summit in London on February 11th. Bandmates Nick Cocking and Hari Ma quickly learned accompanying parts on the mandolin and fiddle, preparing for a reunion two years in the making. “I wheeled him out on stage, and neither of us could believe it was happening,” Cave recounted. “He was thrilled.”
As Darling remained on stage, his AI-generated voice filled the room, accompanied by the live instrumentation of his bandmates. The performance was a powerful testament to resilience, innovation, and the enduring power of music. Cocking and Cave both express Darling’s intention to continue creating music with these tools, though acknowledging the unpredictable nature of ALS makes long-term planning difficult.
“It’s so bittersweet,” Cocking admitted. “But getting up on stage and seeing Patrick there filled me with absolute joy. I know Patrick really enjoyed it as well. We’ve been talking about it … He was really, really proud.”
Frequently Asked Questions About ALS and AI Voice Cloning
What is ALS and how does it affect individuals?
Amyotrophic Lateral Sclerosis (ALS) is a progressive neurodegenerative disease that affects nerve cells in the brain and spinal cord, leading to muscle weakness, paralysis, and eventually, difficulty with breathing and swallowing. It impacts individuals differently, but ultimately leads to a loss of motor function.
How does AI voice cloning work for people with ALS?
AI voice cloning utilizes machine learning algorithms to create a synthetic replica of a person’s voice based on existing audio recordings. For individuals with ALS who are losing their ability to speak, this technology allows them to communicate and express themselves using their own voice, even as their physical capabilities decline.
Is the AI-generated voice perfectly identical to the original?
While AI voice cloning has advanced significantly, the resulting voice isn’t always a perfect match. Often, subtle nuances and imperfections remain, which can actually contribute to the authenticity and emotional impact of the cloned voice. In Patrick Darling’s case, his AI voice retains a slight raspiness from his original recordings.
What are the limitations of AI voice cloning technology?
The quality of the voice clone depends heavily on the amount and quality of the original audio recordings. Creating a singing voice clone is particularly challenging, requiring more data and sophisticated algorithms. Additionally, the technology doesn’t address the underlying physical challenges associated with ALS.
Where can I learn more about ElevenLabs and their impact program?
You can find more information about ElevenLabs and their initiatives to support individuals with voice loss on their website: https://elevenlabs.io/blog/impact-program-announcement. Their impact program offers free licenses to their voice cloning tools for eligible individuals.
Discover more from Archyworldys
Subscribe to get the latest posts sent to your email.
