A Mobile OS that Gives Voice to People with Speech Disorders
October 2024
Isn’t it frustrating when you know exactly what you want to say, but you just can’t make yourself fully understood? What if you had no direct way to assess the results of your efforts and instead had to rely on other people’s reactions?
This is the constant challenge of people with hearing loss. Even if they improve their enunciation with a speech therapist, it’s hard or even impossible for them to judge if what they’re saying is landing correctly. The look in the eyes of the person they’re talking to is the first clue as to whether or not they’ve been understood.
The short video “Find your Voice” illustrates the relevance of the solution to the speech-impaired
Enter Celia Voice Enhancement, a feature of Huawei’s mobile OS HarmonyOS NEXT. Launched in China on October 22, 2024, Celia Voice Enhancement recreates the speech of people with a speaking disorder, which is common among those affected by hearing loss. And not only that, rather that generating a disembodied mechanical voice, Celia emulates the unique timbre and character of the user, accurately expressing a key trait of their unique personality.
In China, 27 million people are deaf or hard-of-hearing. Of those, about 7 million rely on oral communication, but have a speech disorder that’s associated with hearing loss. This makes it difficult or impossible for them to correct their speech.
Globally, about 450 million people, representing 5% of the world population, are affected by hearing loss. By 2050, the World Health Organization predicts that 700 million will be affected.
When someone with a speech disorder first uses Celia Voice Enhancement, all they need to do is say a few sample sentences that enables the AI to individualize the solution just for them. From then on, the user simply speaks into their smartphone when talking to someone face-to-face, and Celia resays the words and sentences in a way that is clear and easily understood by the listener.
A user providing a voice sample to Celia
Celia Voice Enhancement is based on large voice model timbre cloning technology that Huawei developed specifically for people with speech disorders. Dozens of Huawei R&D staff put their hearts into this project for over a year, collecting training data, developing the engine, and implementing the project from start to finish. Project testing started in 2023, with volunteers with hearing loss lending us their time.
One particular challenge during development was the relatively few voice samples in the Chinese language from people with speech disorders. However, the team managed to piece together over 10,000 hours of training data, enabling the technology to reach a high level of accuracy.
Parents with speech disorders don't need to fear teaching unclear speech to their children
Currently, Huawei’s R&D team is still working on Celia Voice Enhancement, continually improving its accuracy to deliver a smoother experience for the people who can benefit most.
