Customer kiosk
24/7 in-store & lobby guidance.
Sona Persona · AI Human
Real-time conversational AI human
No install, no signup. Start a free 5-minute session right in the browser.
How it works
Choose an avatar
Pick from preset personas or train a custom avatar.
Real-time voice & video chat
Talk directly in the browser, no install. It responds in ~1.5s for text and ~2s for voice.
Embed into kiosk, web, or app
Integrate via SDK/embed for 24/7 unattended operation.
Where it's used
Education
1:1 tutoring & language learning.
Healthcare info
Reception & wayfinding in hospitals.
Museum & public info
Exhibition docent & tourism.
Retail
Styling recommendations & in-store guide.
Entertainment
AI character performance & interactive media.
Compared to existing options
| Legacy video production | Live human agent |  Sona Persona | |
|---|---|---|---|
| Cost | Millions per shoot | Salary + 24/7 ops | Usage-based |
| 24/7 availability | Static content | Shifts required | Unlimited |
| Multilingual | Re-shoot required | Additional foreign-language staff | 24 languages supported |
| Connection wait | No interaction | Queue / call wait | Instant, no queue |
| Consistency | Identical | Varies by person | Always identical |
Measured performance
From speech recognition to lip-synced video, one in-house real-time pipeline handles it end to end.
- 1.5-2s
- First response
- ~1.5s text · ~2s voice
- 260ms
- Live transcription
- Transcribed as you speak
- 25fps
- Zero-dropout lip-sync
- No audio dropouts, start to finish
Frequently asked questions
It varies by requirements and integration scope, but standard adoption is possible within 1-2 weeks.
Yes if GPU infrastructure is in place. We support stricter compliance setups for public-sector and finance. Contact us for details.
By default, all session data is discarded at session end. Optional retention/analysis is available on request with secure storage.
Pricing is custom based on scale and integration scope. Submit the inquiry form and we'll reply within 2-3 business days.