Zero spam. Unsubscribe at anytime.
We are surfing the wave of innovation with those who can keep up with the speed. Join over 1000 top investors, doctors, and entrepreneurs in the VoiceMed newsletter to stay updated with our story, our deep tech releases, and product milestones.
Clinically-proven breathing exercises for asthma with sound-based feedback in the Asthma CLIMB program
Monitor the risk of future worsening of your asthma. Feature under development
Leading healthcare organizations, accelerators and universities that are supporting VoiceMed
Voice carries a vast array of information about one's health. The simple act of speaking requires coordinating our brain, lungs and more than 100 muscles. By combining sound analysis with artificial intelligence, we detect subtle signals in the sounds produced by the vocal apparatus. Through a rigorous research process, these signals can become vocal biomarkers.
Our solution is based on two main technological components:
a mobile front-end application that runs on both Android and iOS devices;
a cloud-based back-end application that communicates with the mobile application through a set of REST APIs.
The mobile application provides a convenient way for users to access clinically-validated breathing exercises, educational content such as video or audio recordings, and other features.
The cloud-based back-end application provides modules to be consumed through the mobile application and manages data produced by it. It includes a set of advanced machine learning and sound analysis services that analyze the user’s audio recordings and provide personalized feedback.
Our back-end application is modular in design to ensure ease of extensibility and integration with other systems. Modules are fully containerized and can be run on any cloud platform.
We're a highly-skilled, international, goal-oriented team of superstars. We have people who are crazy about voice, amazing developers, restless scientists, and much more. Our team combines Ph.D. researchers, entrepreneurs and tech experts. Our uniqueness relies on our strong and demonstrated ability to strategize and execute and on our full skillset. Our vision is to empower everyone, everywhere, to know their well-being!
Every day, we produce large amounts of vocal data. Through our voice, we communicate and interact in a simple and direct way. Just look at the growth of voice recordings on social media, the boom in smart speakers like Alexa (175 M in US) and the boom in teleconsulting which could add significant value to the visit. To collect a voice sample, you only need a smartphone - a device that is now widespread all over the world.We produce large amounts of data, but what value can we give it? The world is experiencing a digital transformation and, among all the data we create, voice will play a pivotal role. Now the time has come to give it the right value, especially from a medical point of view, because it is for patients, easy to give and amazingly informative, almost like blood tests.
Voice is related to important physiological activities and is the result of the interaction of different parts of the human body: brain, heart, lungs, and over 100 muscles at the facial level. Its peculiarity lies in the "fine" mechanism that generates it, which makes the voice to be extremely sensitive to changes and imbalances often associated with a pathological state. By studying and analyzing the acoustic characteristics of breathing, coughing, and speech, it is possible to extract relevant information on the evolution of patients' health status. There are four main areas in which voice can play a major role:
Neurodegenerative diseases: speech analysis, even remotely, allows the extraction of indicators whose purpose is to monitor the patient and provide further information regarding a possible change in therapy, allowing to notify the doctor beforehand of the booked visit.Neurodegeneration, as in ALS patients, can fastly impair a patient's mobility. Thus, improving remote monitoring could of great help. Also, several studies show how vocal biomarkers are now being explored in the early diagnosis of Parkinson’s and Alzheimer’s disease.
Respiratory diseases: because, in the case of breath and cough, the acoustic activity of the subject is directly related to the airflow, it is possible to gather interesting information about the airways. For instance, in the case of chronic obstructive pulmonary disease (COPD), dyspnea, or asthma, information about can be obtained checking both the typical obstructive/restrictive and the generalfluid dynamics patterns.
Mood disorders: vocal communication can be affected by many factors, including psychological status. The vocal dynamicsinherent in tone, spelling, pauses, and the whole dimension of speech areobjects of analysis as a further indicators of situations of unease, euphoria, insecurity, and much more, constellating a large and wide domain capable of providing a valid point of view regarding the mental and the psychosomatic state of a subject.
Cardiovascular diseases: the vocal production system, having as a nucleus the vocal cords, is supplied by a complex system of vascularisation that balances its behaviour to muscle stimulation and fluid dynamics. Cardio-related anomalies can ultimately result in unbalancing the whole system, which therefore takes on a new information content transmitted to the acoustic production.
A biomarker is a characteristic that indicates normal biological and pathogenic processes that may include molecular, biochemical, histological, radiographic, or physiologic characteristics. With the digitalisation of healthcare, the definition of digital biomarkers has emerged: “objective, quantifiable, physiological, and behavioural measures that are collected through digital devices” that are “portable, wearable, implantable, or digestible” (Babrak et al. 2019). Though a formal definition of vocal biomarkers is yet to be framed, we define vocal biomarkers as a specific feature of speech, breath sounds and cough that indicates the underlying physiological and/or pathophysiological condition. To be considered valid, such biomarkers, as any digital biomarker, must have their performance and clinical significance rigorously assessed.Vocal biomarkers are observable as the result of complex and refined analysis of a generic human acoustic production. They can compress and summarise the pathological information contained within it. At the same time, they have the property of generalisation and specificity, thanks to the enormous amount of data on which they are constructed and calibrated both on a global level (inter-subject analysis), and on a personal level (on the basis of the nature of the individual subject by intra-subject analysis). The speech biomarker study has three main components: medical, engineering, and development. In fact, the companies that study vocal biomarkers are multi-competent groups, such as VoiceMed: they include the medical and data collection part in clinical contexts, then they perform sound analysis through artificial intelligence, and finally the product part, to ensure that the way in which data is collected is in line with the use of the device in real contexts and with regulatory bodies, for the approval of the future medical device.