Museum audio description communicates visual and spatial information that is relevant to accessing an object, image, environment, or action. Good description is selective, structured, precise, and developed with blind and low-vision people—not a verbal inventory of everything in view.
The practical sequence: orient the visitor, establish the whole, locate important parts, describe the evidence that supports interpretation, distinguish observation from interpretation, and connect the description with touch, navigation, and choice where available.
1. Define the purpose and context
Decide what the description must make accessible. Is it an independent visual-description route, part of the main guide, a description of video, an orientation to architecture, or support for a tactile encounter? Record viewing distance, physical route, nearby sounds, touch opportunities, and safety information.
2. Orient before detail
Start with identity, scale, format, position, and overall composition where relevant. Give a stable frame before moving through detail. For a painting, that might mean dimensions, orientation, dominant arrangement, and the visitor’s location relative to it.
Use a consistent spatial system: left to right, foreground to background, center outward, or clock-face positions. Announce the system instead of switching unpredictably.
3. Select meaningful visual evidence
Description cannot include everything. Prioritize information needed for orientation, the work’s interpretive purpose, distinctive form, material, gesture, color, condition, technique, and relationships. Include details sighted visitors are being invited to use.
Selection is an editorial responsibility. Consult curators and access specialists, then test what users find useful rather than assuming sighted priorities are universal.
4. Use precise, economical language
Name rather than hedge when evidence supports it. Use concrete nouns and verbs, explain unusual visual terms, and avoid relying on color alone. Comparisons can communicate scale or form if they are culturally understandable and do not introduce more complexity.
- Avoid “obviously,” “simply,” and “as you can see.”
- Do not romanticize blindness or address the visitor as a passive recipient.
- Use pauses so details can be processed or located.
- Check names, materials, directions, and pronunciation.
5. Separate description and interpretation
Objective description is never perfectly neutral: selection and language shape meaning. Still, signal the difference between observable evidence, documented fact, and interpretation.
Less useful: “A tragic queen looks hopelessly toward her fate.”
More transparent: “At the center, the seated woman lowers her head and clasps both hands. Her face is turned away from the open doorway. The museum interprets this withdrawn pose as anticipation of the sentence to come.”
6. Integrate touch, navigation, and other access
Connect description with tactile models, replicas, raised diagrams, material samples, orientation maps, and facilitated touch where appropriate. Explain where an element is, how to approach it, what can be touched, and what cannot.
Provide transcripts and accessible controls. Description does not replace captions, sign language, readable text, physical access, or staff support.
7. Review and test
Use blind and low-vision consultants as paid experts and test in the real location. Review orientation, mental model, relevance, pace, language, navigation, fatigue, audio quality, and whether description leaves room for response.
AI can structure a draft from controlled data but cannot reliably decide what is meaningful or appropriate in context. Keep expert review and publication control.
8. Where GuideSofia fits
GuideSofia can publish description as a dedicated route or integrated option, connect it to transcripts and tactile stops, and maintain versions across languages. The institution determines the object knowledge and access approach.
9. A compact writing example
Orientation: “This vertical oil painting is about the height of an adult and half as wide.”
Whole: “A single figure stands against a pale wall, occupying almost the full height of the canvas.”
Evidence: “The coat is painted in broad, dark strokes. Both hands grip a folded letter at waist level; the paper is the brightest shape in the picture.”
Interpretive connection: “That contrast makes the letter the visual center of the scene. Archival correspondence identifies it as the notice that ended the subject’s employment.”
10. Audio-description checklist
- Purpose and physical context are defined.
- The visitor receives orientation before detail.
- Spatial language is consistent.
- Selected details support access and interpretation.
- Observation, fact, and interpretation are distinguishable.
- Language is precise, respectful, and speakable.
- Touch and navigation are integrated where relevant.
- Blind and low-vision experts review and test the final experience.
Frequently asked questions
What is museum audio description?
It is spoken language that communicates visual and spatial information relevant to accessing an object, image, environment, action, or interpretive experience, primarily for blind and low-vision visitors.
Should audio description include interpretation?
It may include clearly framed interpretation, but should distinguish observable evidence from interpretation so visitors can understand what is present and form their own responses.
Can AI generate museum audio description?
AI may assist with structured drafts, but it cannot determine relevance, cultural sensitivity, physical context, or user usefulness reliably on its own. Expert and user review remain necessary.
Is alt text the same as audio description?
No. Alt text is usually concise and serves a specific digital context. Museum audio description may provide structured orientation, detail, spatial relationships, material qualities, and interpretive context over a longer listening experience.