Johns Hopkins University defines stuttering as a speech-language disorder. Stuttering affects more than 80 million people around the world, and in the United States more than 1 million Americans have a stutter.
Vocal disorders are problems with the pitch, volume, timbre, or other characteristics of the voice that occur when the vocal cords do not vibrate properly. There are several types of voice disorders that are classified as organic, including structural and neurological disorders (caused by neurological disorders such as Parkinson's disease and Alzheimer's disease), functional disorders (muscle dysfunction), and psychogenic disorders. Includes sexual disorders.
A Dutch startup has developed an app designed to provide a voice to people with speech and language disorders such as stuttering.
The Whispp app uses artificial intelligence (AI) to make phone and video calls clear and relaxing.
Joris Castermans, CEO of Whispp, said: “This app has the ability to convert speechless/impaired vocal cords and whispered speech into natural voiced speech. It is equipped with real-time voice-assistance technology. “For example, people who severely stutter can reduce their stuttering frequency by an average of 85% when whispering. Additionally, people who suffer from spasmodic dysphonia or recurrent respiratory papillomatosis can reduce their stuttering frequency by an average of 85%. People who speak in a whisper are much more relaxed and fluent when speaking in a whisper.”
In an email interview, Castermans said Whispp allows users to express themselves better and more easily, improving their quality of life and allowing them to participate more fully in society. .
“Communication is a fundamental aspect of human existence and is a daily challenge for people who suffer from voice disorders and severe stuttering,” Castermans said. “Lack of communication skills can lead to social isolation, which can often lead to feelings of inadequacy and depression.”
In its proprietary AI model for Whispp, the company does not use a language model as the AI is speech-to-speech based with no text intermediates.
“This allows Whisper to convert non-voice audio with very low latency,” Castermans said. “Whispp’s AI converts every 20 milliseconds of audio into a real-time stream.
Whispp uses real-time voice-to-speech-based auxiliary voice AI to create real-time voice translation to accommodate a wide variety of voice types, from whispers to gullet rasps. This allows the app to create solutions tailored to different audio conditions.
“Communication is a fundamental aspect of human existence and is a daily challenge for people who suffer from voice disorders and severe stuttering,” Castermans said. “Lack of communication skills can lead to social isolation, which can often lead to feelings of inadequacy and depression.”
For example, Castermans notes that people who severely stutter speak fluently and are relaxed when whispering. “This is due to neurological changes that occur while speaking. Separately, people with severe stutters did not 'learn' to become anxious while whispering.”
Castermans said major tech and voice-assisted technology companies are primarily focused on automatic speech recognition, known as speech-to-text (STT) for non-standard speech. “This is extremely useful for patients with impaired articulation (ALS, MS, stroke, Parkinson’s disease) who can synthesize speech using text-to-speech.”
“However, the downside to this approach is the high latency of two to three seconds, creating a barrier to natural conversation,” Castermans says. “As a result, current AI voice technology solutions cannot provide a suitable solution for people with voice disorders who have lost their voice but still have good articulation.”
Whispp app is available for Android and IoS.
follow me twitter.
