Voice interface
Connecting audio, conversation states and a display that shows the user what is happening.
Available test version · active development
Voice and text in the same conversation.
A conversational experience connecting voice, text and user context. Users switch between modes, provide personal instructions and control memory.
Open SonimaConversation and memory views use demonstration data.
Actual view of the public Sonima page.
Interface with demonstration data: continuing a voice conversation in writing.
Interface with demonstration data: personal instructions and memory controls.
User experience
A conversation is more than a model response. Starting a voice connection, listening states, returning to text and understandable context controls all matter.
After a voice conversation, the user can continue in writing within the same conversation. The interface shows the transition so that it is clear where the conversation continues.
Users can specify how they want the assistant to respond. Instructions are edited separately from remembered notes, so conversational style and memory are distinct settings.
Users can review, correct or remove remembered information. A temporary-conversation option is also available, with memory status visible in the interface.
What this development connects
Connecting audio, conversation states and a display that shows the user what is happening.
Separating user instructions, previous messages and memory, with their own controls.
A consistent flow between voice, text and conversation review across screen sizes.
Our own development
An AI platform for patients and physicians: conversation, search and specialist sources.
A Windows tool for visual mapping, validation and transformation of HL7 and JSON data.
Planned coding suggestions with source highlights and professional review.
Let’s connect
We can help define the workflow, voice or text interface and how context connects with your application.