IstoVisio · B2B · Voice · AI
Redefined Task
Execution with an
AI Voice Interface
syGlass · January 2025
syGlass is a VR-based scientific visualization platform used by scientists to explore and analyze massive 3D datasets. As workflows became more complex, the number of interaction steps required to complete common actions increased significantly, breaking focus during analysis.
This project focused on reducing the time between user intent and system action through an AI-powered voice interface, allowing scientists to execute commands more efficiently while maintaining focus during analysis.
Problem
Scientists knew exactly what they wanted to do, but executing those actions required navigating multiple panels, configuring settings, and switching between tools. These repetitive interactions interrupted focus and slowed analysis during high-frequency workflows.
As workflows became more complex, the cost of these interruptions increased. The business needed a faster way for scientists to execute common actions without sacrificing precision or disrupting existing workflows.
Solution
I designed an AI-powered voice command system that translated spoken instructions into system actions.
The redesign included:
- • Designing voice as a workflow accelerator rather than a replacement for existing interactions
- • Supporting natural language variations to accommodate different accents, phrasing, and speaking styles
- • Creating clear recovery paths when commands were misunderstood to help users stay in control
Impact
The final solution reduced the time required to complete common workflows by 41%.
The redesign also:
- • Reduced the number of interaction steps required to perform common tasks
- • Improved efficiency for expert users during complex analysis sessions
- • Established a foundation for future AI-powered interactions within the platform
My role
As the founding designer, I:
- • Led the project end to end across research, voice UX, interaction design, prototyping, and validation
- • Defined the voice interaction model, command structure, and error recovery framework in partnership with engineering
- • Designed onboarding, discoverability, and trust mechanisms to support adoption within existing scientific workflows
To understand where the platform slowed scientists down, I conducted interviews and observational sessions with eight scientists across Europe and the United States. I focused on how users moved through analysis workflows and where the interface interrupted their momentum.
“The software isn't difficult to use, but I constantly have to stop what I'm doing to make small adjustments.”
— Scientist, syGlass user research participant
Three patterns emerged:
Navigation Overhead
Completing even simple actions required moving through multiple panels, menus, and settings before users could make a change.
Context Switching
Routine adjustments such as changing opacity, toggling visibility, or modifying settings repeatedly interrupted analysis workflows.
Discoverability
Frequently used commands were buried within the interface, making them difficult to find and requiring repeated navigation to access.
Research showed that workflow friction came from the effort required to perform actions within the interface. Scientists needed a faster path from intent to action while maintaining focus during analysis.
Several approaches could have reduced interaction cost, including shortcuts and additional interface controls.
Voice was selected because it allowed actions to be performed from within the workflow itself, without introducing additional interface complexity. It complemented existing controls, preserved user choice, and scaled across a wide range of scientific tasks without requiring users to learn a new navigation model.
Across two rounds of remote usability testing with eight scientists, participants quickly understood the value of voice commands but struggled to remember available commands and exact phrasing.
Adoption depended on making commands easy to discover and learn over time.
To support adoption, I designed an AI-powered voice interaction model that reduced the number of steps between intent and execution while making commands easy to discover through natural language interactions.
Scientists could perform actions directly from their workflow using natural language, while continuing to rely on existing controls whenever needed.
For voice to become part of everyday scientific workflows, the experience needed to be easy to understand, reliable across different speaking styles, and flexible enough to fit existing ways of working.
Barrier
Diverse Speech Patterns
Scientists expressed the same intent using different accents, phrasing, and speaking styles.
Design Response
- Multiple command variations
- Flexible phrase recognition
- Recovery prompts
Barrier
Learnability
Voice interactions needed to be easy to discover and use over time without relying on memorization.
Design Response
- Categorized commands
- On-demand command menu
- Tutorials and guidance
Barrier
Workflow Compatibility
Scientists had established workflows and interaction habits that could not be disrupted.
Design Response
- Hybrid voice and menu interaction model
- Voice as an optional workflow accelerator
Scientists often expressed the same intent in different ways depending on their vocabulary, speaking style, accent, or level of English fluency. Requiring users to memorize specific commands would have created unnecessary friction and limited adoption.
To make voice interactions feel natural, I designed the system around intent recognition rather than exact phrase matching. Each command supported multiple natural variations, allowing scientists to speak in the way that felt most intuitive to them while still producing predictable outcomes.
Increase performance
Expected Inputs
Decrease performance
Expected Inputs
Increase opacity
Expected Inputs
Voice commands only become useful when users can remember how to use them. Through research, it became clear that scientists would not adopt the feature if they had to memorize a large library of commands before seeing value.
To reduce the learning curve, I organized commands into predictable patterns and supported them with in-product guidance, tutorials, and documentation. The consistent structure helped scientists build familiarity with the system over time while allowing new users to quickly understand how commands were constructed.
Rather than expecting users to learn the system, the experience was designed to help users learn naturally through repeated use.
Voice interactions introduce uncertainty. Users need to understand when the system is listening, what it recognized, and how to recover when a command is misunderstood.
To support this, I designed a feedback and recovery system that made the system's state visible and provided clear paths forward when interactions were unsuccessful.
The interaction model was designed to support graceful recovery. When commands were incomplete or partially understood, the system could clarify intent, suggest next steps, and help users continue their workflow without starting over.
Visual feedback made the system's state visible, giving users confidence that their voice had been detected and reducing repeated commands.
Recovery prompts guided users toward successful outcomes, helping them stay confident in the interaction without interrupting their workflow.
The final experience shortened the distance between intent and execution while preserving the precision and control scientists relied on during analysis.
By working alongside existing interactions, voice provided a faster path to action without requiring users to change the way they worked.
The final solution reduced the time required to complete common workflows by 41%.
The redesign also:
- • Reduced the number of interaction steps required to perform common tasks
- • Improved efficiency for expert users during complex analysis sessions
- • Established a foundation for future AI-powered interactions within the platform
The most successful products do not force users to choose between efficiency and familiarity. They create new capabilities while preserving the behaviors people already trust.
This project reinforced the value of designing around existing workflows and demonstrated how adoption depends on making change feel incremental rather than disruptive.
