Australian start-up Laronix is developing a bionic device that uses AI-based voice-cloning technology to restore the voices of laryngectomy patients.
Dr. Farzaneh Ahmadi received an email ten years ago while researching for her PhD in Singapore.
She was contacted by a laryngeal cancer patient in the United States while working on a team exploring bionic voice technology.
“A patient emailed our group and said, ‘I’m facing this condition; I will lose my voice and I’m a singer — my voice is my life. Can you help me?’”
Ahmadi was disappointed that the research had not progressed far enough to restore the patient's original voice.
“Our honest answer was, maybe in 20 years,” she states.
“At that point, it started to bother me personally. Why can’t we solve this — something that seems so biologically possible?”
Old-school solutions
The voice box is removed during a laryngectomy, and the trachea (windpipe) is relocated to the front of the neck, where a stoma is formed to allow breathing.
Because each patient's anatomy is different, so does voice rehabilitation following surgery.
“There is huge variation [among patients],” says Alanna Bowen, a Laronix-affiliated speech pathologist.
“It’s individualised and it’s purely down to how people heal from surgery, the amount of treatment they need to get, their own capabilities, dexterity, support people. It’s a patient population that has to be treated individually because there’s just no one size fits all.”
Existing options include vocal prosthesis and the electrolarynx, a robotic-sounding device that activates when held to the neck. According to Bowen, while many laryngectomees have success with these, some have issues.
A one-way silicone valve is implanted into the tissue between a patient's windpipe and oesophagus in the tracheoesophageal voice prosthesis, which is considered the current gold standard. This device sends air to the back of the neck, which vibrates and produces sound, when a patient covers their stoma.
Although it can generate a powerful voice, it must be cleaned and changed on a regular basis, which can be problematic for some people.
“When I was working clinically for public health, I would have patients travel 500 km to come and see me for a voice prosthesis change,” Bowen adds.
“I was the closest person who had the skill to be able to give the service that they needed.”
Her grandfatherhad a laryngectomy when she was studying. He utilized an electrolarynx after testing a prosthetic.
“[My grandparents] had a business in town and it took a long time for people [to adjust]. Even though they knew my grandfather, they knew he’d had this surgery, and it was a little country town — if he would pick the phone up for the business, people would hang up.”
The location and complicated construction of the larynx, according to Ahmadi, is one reason for this.
An unexpected find
Ahmadi was born in Tehran and worked as a telecom engineer before moving to Australia through Singapore in 2013.
She walked away from bionic voice research after becoming frustrated with the lack of progress.
When she applied for a grant to study optimality in biology instead, she was turned down, which turned out to be an unexpected source of inspiration.
“One of the reviewers had said, ‘If biology is so optimal, why aren’t you using that principle in generating voice for people, which has not worked in your PhD?’ That was, ironically, a eureka moment,” Ahmadi says.
She went on to the University of Sydney after that. During her lengthy literature analysis, she came across research from China that described a pneumatic artificial larynx (PAL), which is a very rudimentary mechanical larynx that creates voice using a person's breathing impulses.
“It’s the world’s simplest model of human larynx — it’s out of the body, completely mechanical,” Ahmadi explains.
“Ironically, it generates Chinese language, which is a tonal language, meaning that it can generate pitch variations inside the word, something that healthy people also find difficult to do.”
This was a crucial find.
“It was the ultimate proof in our hand — that the only thing that you need to convey an exceptionally high-quality voice for humans is respiration; you don’t need any commands from the brain. That immediately simplified the problem,” Ahmadi states.
By 2018, Ahmadi and a team at the University of Western Sydney had developed a Pneumatic Bionic Voice based on the PAL that had gone through pre-clinical testing.
Laronix was formed by Ahmadi and her brother, Dr Mousa Ahmadi, as a spin-out from the University of Western Sydney in 2019.
Bionics Queensland's Bionics Challenge, which aims to stimulate bionics development in Australia, awarded the start-up a major category last year.
Dr. Robyn Stokes, Director of Bionics Queensland, was impressed by Laronix's thorough research and development.
“Where it stands out is the quality,” she says. “The clarity and the non-invasive aspect of it and possibly how far [they’ve] gotten in terms of tapping what that artificial intelligence can do in that context — it’s much more advanced [than other solutions].”
According to Stokes, the technology could also have less obvious applications.
“It has uses beyond for those with laryngectomy. [For example,] there are millions of dollars lost economically from the loss of voice in the education sector.”
AI Assistance
Bionic Voice, unlike surgical alternatives, is a non-invasive electronic gadget that monitors signals generated when a user breathes.
This data is transferred to an AI module housed in a headgear that mimics vocal chords.
The voice is then conveyed to the mouth via a thin tube in a recent model, but Laronix wants to streamline the design even more. According to Ahmadi, the ultimate goal is to restore people's natural-sounding voices.
“Once you generate voice, the second step is to train that voice to become more natural. That’s where the AI kicks in,” she explains.
The AI program is fed audio recordings and can be taught to replicate a certain voice in five to 20 minutes, with quality improving over time.
Users will be able to change the tone of their voice, sing and laugh, and one day, thanks to artificial intelligence, they may even sound as they did before surgery.
Acceleration
According to Ahmadi, getting a product from a university to the market hasn't always been easy. However, the company has received significant public and private funding, and trials will begin this year.
“She completely understands the global marketplace”. says Stokes of Ahmadi's leadership.
Ahmadi is still determined to assist as many people as he can.
“For every one of our patients that gets their voice back, you’re just ultimately so thrilled,” she says. “At some point, it becomes a matter of numbers – how many lives you can save.”
Finding a voice
Bobbi, a laryngeal cancer survivor, was one of the first patients to try the Bionic Voice (last name withheld for privacy).
She was previously using a voice prosthesis, but the valve grew loose and began to leak due to tissue damage in her neck caused by radiation, causing major difficulties.
Bobbi had been silent for several years before working with Laronix, which caused her everyday difficulties and had a significant impact on her mental health.
“I use [an e-writer] or a text-to-speech app on my phone, but both forms of communication are so much slower than verbal communication that by the time I have written my thoughts the conversation has moved on and what I have written is no longer relevant,” she adds via email.
Bobbi believes she can "be part of society again" using the Bionic Voice.“Socialisation would be easily possible; parties and meetings and groups would be accessible. My mood would lift enormously,” she adds.
“I could put the device on and order coffee verbally, I could go to the butcher and ask for items, I could tell my daughter she looks lovely today, I could yell at my grandson, I could train my dog, I could go to dinner with friends and participate in the conversation. I could be/feel normal again.”