Why AI native matters
Other AI approaches either reset at each painting or skip guidance entirely. Musa does both: structured tours with continuous conversation.
Guided paths, not blank canvases
Visitors don't need to ask 'what am I looking at?' Curators design structured tours while AI handles the delivery. Press next or dive deeper. Your choice.
Continuous conversations
Unlike chatbots that reset at each painting, Musa maintains context across your entire visit. Reference something from three stops ago. It remembers.
Spatially aware by default
Musa tracks where visitors are and adapts scripts, answers, and directions automatically. No guessing, no QR-code maze.
Museum-controlled knowledge base
Curators stay in charge of every fact, source, and prompt. AI responds within guardrails you define. Publish updates instantly.
The evolution of audio guides
From hardware to AI-native: each generation solved one problem but left others unsolved.
From source material to a live AI audio guide
The path from your archive to visitors' phones is short, and your curators hold the pen at every step.
Upload source material
Wall texts, catalogues, PDFs and media become a structured knowledge base that only your guide can use.
Curators review and set guardrails
Your team shapes tours, tone and personas in Studio, and defines what the AI may answer and how supplemental context is labelled.
Launch by QR code
Visitors scan and the guide opens in the browser on their own phone, in 40+ languages, with no rental hardware.
Read the questions, improve the guide
Analytics show what visitors ask at every stop. Update content in minutes, and the next visitor hears the new version.
The AI capability plugs into the rest of the platform: the audio guide software workspace, PDF and catalogue import, the creation workflow, multilingual delivery, and spoken visitor navigation.
FAQs about AI audio guides
We combine guided tours with live conversation. Visitors follow structured paths designed by curators, but can ask questions anytime, and the AI remembers context across their entire visit, not just per painting.
Pure Q&A forces visitors to prompt from scratch at every stop. 'What am I looking at?' gets old fast. Guided tours give structure. Press next for the curated story, or dive deeper when curious. Best of both worlds.
Those chatbots reset at each painting. Ask about the artist's technique at stop 3, and by stop 7 it's forgotten. Musa maintains a continuous conversation. Reference earlier stops, build on previous answers, create a coherent journey.
Yes. Visitors use their own phones. We handle payments, timed access, and usage analytics so you can phase out hardware rentals while maintaining or growing revenue.
Most museums ship a pilot in a few weeks. You can import wall labels, PDFs, and media, then iterate in the Studio without waiting on external production cycles.
Musa is priced as one product, with no separate AI surcharge: revenue share with no upfront costs, a subscription from £500 per month, or prepaid credits from £1,000. Onboarding and content ingestion are included.
Visitor interactions are anonymous and aggregated, so you learn what visitors ask without tracking individuals. The knowledge base stays scoped to your exhibition; there is no internet scraping.
Tour audio has offline support and keeps playing in dead zones. Live AI answers need a connection: they pause while a phone is offline and resume when it returns.
You do. Your curators maintain it, and you can export the full knowledge graph, transcripts and analytics at any time.
The story behind audio guides
From 1952 radio broadcasts to AI-powered conversations. Discover how museums learned to talk to visitors, and why it took 70 years to fulfill the original vision.
Read the full history→Ready to explore?
See how museums deploy Musa or talk to the team about your collection.