How to Choose a Narrator for Exhibitions, Museums, and Audio Guides: Voice Design for Acoustics, Visitor Pace, and Multilingual Delivery

A Great Voice Alone Is Not Enough for Exhibition Audio Guides

The most common mistake in exhibitions, museums, and galleries is choosing a narrator based only on a pleasant voice. A voice that works beautifully in a video may fail inside a physical exhibition space. Visitors are not seated and fully focused. They are walking, pausing, turning, and listening through ambient chatter, footsteps, and reflected sound.

So the first criterion is not vocal charm, but intelligibility. In practice, pay attention to consonant clarity, tail-end decay, breath noise, and the information density of each sentence. In reverberant spaces, a warm low-end voice often loses detail, while a voice with controlled presence around 2kHz to 4kHz tends to cut through better. In many venues, crisp consonants matter more than lush resonance.

Choosing a Narrator with Room Acoustics in Mind

In exhibition environments, ambient sound often sits around 55 to 70 dB. Glass cases, stone walls, and atriums increase early reflections and blur word onsets and endings. That is why pre-audition listening tests are essential.

I recommend checking narrator samples under three conditions:

1. Smartphone built-in speaker
2. Bone-conduction or open-ear headphones
3. Audio processed with simulated venue reverb

You can create useful simulations with iZotope RX, Adobe Audition, Waves IR-L, or Altiverb. If the expected RT60 exceeds about 1.2 seconds, a stable read with limited pitch swings is usually easier to understand than an expressive, highly dynamic delivery. In contrast, a quiet gallery using personal devices may benefit from a narrator who can convey subtle emotion. The key point is simple: the right voice changes with the space.

Start from Visitor Walking Speed, Not from the Script

Audio guides should be designed from visitor flow, not from copy length. In many venues, average visitor walking speed falls around 50 to 70 meters per minute. Average dwell time at one exhibit is often 20 to 45 seconds for standard displays and 60 to 90 seconds for highlighted works.

Using those conditions, practical narration lengths per stop are roughly:

• Pass-through displays: 20 to 35 seconds
• Standard displays: 35 to 60 seconds
• Featured displays: 60 to 100 seconds

For Japanese, a useful benchmark is 240 to 300 characters per minute for audio guide narration. In venues with many senior visitors or multilingual device use, slowing to 220 to 260 characters per minute is safer. Do not optimize for maximum information. Optimize for one-pass comprehension while walking.

In Multilingual Production, Align the Role of the Voice Before Translation

A frequent problem is tonal mismatch between languages. The Japanese version may sound calm and scholarly, while the English version sounds bright and promotional. Even when the content matches, the overall brand experience becomes inconsistent.

For multilingual projects, native fluency is only one part of the equation. You must also define the function of the voice consistently across languages. At minimum, specify these four items for every language:

• Relationship to the visitor: guide / curator / storyteller
• Tempo: contemplative / circulation-driven
• Emotional level: restrained / moderate / rich
• Sentence endings: conclusive / lingering

English can sometimes run shorter than Japanese for the same meaning, but pronunciation checks for names, titles, and historical terms often offset that advantage. In real projects, planning English at about 85 to 100 percent of the Japanese duration is a safe approach. If Chinese or Korean is added, reserve an extra 3 to 5 seconds per stop for device switching and operational buffer.

How to Give Better Audition Instructions

For exhibition work, the audition should not mainly test dramatic skill. It should test repeatable information delivery. I suggest asking for three short reads:

• Artwork explanation: fact-centered
• Route guidance: meant for walking listeners
• Visitor notice: calm but unmistakable

These reveal vocal temperature, structural clarity, ending control, and how the narrator applies pressure without sounding harsh. When evaluating, avoid vague comments like “I like this voice.” Use a score sheet with:

1. Consonant clarity
2. Suitability for listening while walking
3. Low fatigue over long listening periods
4. Stability with proper nouns
5. Consistency with multilingual versions

That keeps the selection process practical rather than emotional.

Final Thought: Success Depends More on Design Than on Pure Performance

Audio guides for exhibitions, museums, and galleries are not just narration. They are part of the spatial experience. So the real question is not “Who sounds the best?” but “Whose voice is optimized for this space, this route, and these visitors?”

When I assess a project, I do not start with a flashy demo reel. I start with five conditions: room reverberation, playback device, visitor age range, average dwell time per stop, and number of languages. Once those are fixed, the ideal voice becomes surprisingly clear.

A good audio guide does more than explain objects. It expresses the dignity and identity of the institution itself. So the voice you want is not merely a good voice. It is a voice that deepens the exhibition experience without disturbing it.