Human communication has long depended on speech, but what if words were no longer necessary? AlterEgo, described as the world’s first near-telepathic wearable, offers a silent alternative. Using a technology known as Silent Sense, it captures internal signals associated with unspoken words, enabling soundless interaction with computers and other devices. The system is non-invasive and registers only the thoughts a user consciously chooses to articulate, without relying on audible speech.
Redefining communication via silent sense
In a September 9 update, AlterEgo demonstrated its latest wearable device, which allowed the user to communicate almost telepathically, without making a sound.
In the video, Arnav Kapur, CEO and co-founder of Alterego, was able to set a reminder on his phone, engage in real-time translated conversations in different languages, observe his surroundings and direct silent queries about them to the AI assistant, all without making a sound.
The development of the Alterego wearable is part of an effort to make communication more accessible, particularly for individuals with speech impairments. It seeks to provide an alternative to traditional communication devices, which can be slow and difficult to use. By enabling real-time, silent interaction, the technology may help address some of the limitations faced by non-speaking users.
Why voice input is not enough
With 154.3 million users in the U.S., voice assistants like Siri and Alexa have integrated conversational AI into daily life. However, their usefulness is limited to spoken words, and their accuracy may decline due to various factors.
For instance, their word error rate is 32.5% in noisy environments, compared to only 3% in quiet conditions, highlighting their difficulty in recognizing speech in busy settings. Another limitation is their poor accessibility for users with speech impairments such as dysarthria, a speech disorder characterized by difficulty in articulating words due to weakened muscles for speaking.
Lastly, privacy and security concerns remain significant when using AI-based communication. Research has documented several issues with voice assistants, including accidental activation, unintended recording, third-party permission abuse, and data misuse. These problems can lead to financial, emotional, or privacy-related harm and have often resulted in legal action, such as a $95 million class-action settlement involving Apple’s Siri® voice assistant. Even in the absence of lawsuits, cases like Amazon mistakenly sending 1,700 Alexa recordings to the wrong user highlight the risks of data exposure.
MIT media lab’s Alterego project
Alterego began in 2018 as a project at the MIT Media Lab, an interdisciplinary research laboratory known for supporting unconventional approaches to societal challenges. Under the guidance of MIT Professor Patricia Maes, Arnav Kapur spearheaded the development of the device within the lab’s Fluid Interfaces Group.
Initially, the project was an early prototype designed to detect the user’s inner voice and translate it into digital commands, allowing silent interaction with a computer. Early trials achieved a 92% recognition accuracy rate, demonstrating the system’s potential for reliable, silent communication. Over time, the team improved the device by incorporating AI and machine learning, which expanded its functions and improved its accuracy and responsiveness.
By 2025, Alterego officially transitioned from an academic research project to a commercial venture, joining other startups that originated at the MIT Media Lab. This move aims to translate the research into practical applications and make silent speech technology more widely accessible.
The patent behind Alterego’s silent sense technology
Tracing its origins as an MIT project, Alterego’s patent, U.S. Patent No. 10,878,818, was filed by Shreyas Kapur and assigned to Massachusetts Institute Of Technology (MIT) on September 8, 2018. The patent was granted on December 29, 2020, and the ownership was officially transferred to Alterego AI, Inc. on April 21, 2025.
The patent describes AlterEgo’s silent speech interface (SSI), a wearable device that reads neuromuscular signals generated when a user intentionally speaks words internally. Unlike simply thinking or speaking aloud, silent speech requires consciously forming words inside the mouth and throat, causing tiny movements in the speech muscles without producing audible sound. The SSI captures these signals from the internal speech organs and interprets them to reconstruct the user’s intended words.
To capture these signals, the device uses electrodes, small sensors placed across various facial and neck regions. As illustrated, some electrodes are positioned along the chin, lips, and jaw, while others extend across the lips and cheeks. In Figure 7, electrodes are located under the chin and jaw, and in Figure 8, they are placed on the front and sides of the neck near the hyoid bone and carotid area.
These electrode configurations capture muscle activity from internal speech from multiple angles. The sensors detect the faint neuromuscular signals and transmit them to the SSI for processing. To enhance both accuracy and comfort, electrodes may be made from materials such as silver or other conductive metals.
Lastly, the system employs bone conduction to deliver real-time feedback, allowing users to receive auditory responses silently through vibrations. This creates a closed-loop system where the user silently “speaks”, the SSI decodes their intended words, and then provides a private response. By translating the inner voice into digital commands, the device allows silent, hands-free communication with computers and other devices. This could be useful for people with speech impairments, for private conversations, or for controlling technology without speaking out loud.
This patent, titled “Methods and apparatus for silent speech interface”, lists CEO and co-founder, Arnav Kapur; Shreyas Kapur, and Patricia Maes as inventors. The patent was represented by Stepehen Otis.
Shaping tomorrow in silence
Alterego’s shift from the lab to the marketplace marks only the beginning of its journey. At the recent Axios AI+ Summit, Arnav Kapur shared that the team is refining the final Alterego wearable with a smaller, more compact design. With its non-invasive design built on external sensors, the company envisions a future where silent, seamless communication could become part of daily life.
While Alterego advances along this path, other approaches are also emerging in parallel. Neuralink is testing implant-based interfaces in FDA-approved human trials, while Paradromics is advancing its Connexus® system, which we covered in an earlier article. Together, these efforts highlight the diverse strategies shaping the future of human–computer interaction.
For more on AI-driven communication technologies, see our articles on SeamlessM4T: How Meta’s AI is redefining translation and How Deepgram patents are shaping voice AI
Note: Thumbnail image is for illustrative purposes only and does not represent the actual product made by AlterEgo
