You might find recording your own voice daunting. We did when we started doing it, but because VoiceMap is a personal, intimate medium, we think listeners respond better to you than a trained voice artist. It's your story, after all, and you really can tell it best. Recording is also a lot of fun, once you get going.

People love David Attenborough because he sounds like a grandfather telling an engrossing bedtime story. While that might be an impossibly high bar to set, it's helpful to imagine yourself reading for somebody who knows you, and shares your enthusiasm. You could invite them to give you a hand, if you think that might help.

Below is the final audio of the first location from Derek Spillane's Galway City Centre tour. We'll show you how Derek went from recording his initial sample to the final product.

**Example of final audio**

We've made the process simple and straightforward so you can get the best recording results without all the hassles. Below is an outline of the process once you are ready to record.

1. Your editor will get in touch with you to request a sample.
2. Once your sample is received, we'll provide feedback on how to improve the audio quality and delivery, and request a second sample if necessary.
3. After receiving a second sample, you'll either get the go ahead to record the entire script, or we might request an additional sample, if absolutely required. Usually two samples are enough to get the recording in a good place.
4. Send your recordings of the script to us. We'll edit and mix it for you, then upload it to the tour.
5. Download the tour with the final audio and test it out!

In the below example, Derek has recorded his initial sample in a room without any proper padding and soft surfaces, or setting up a blanket tent as we recommend. You can clearly hear his voice reverberating in the space. He's made a few mistakes, but has repeated the phrases again, so we can edit them together later.

**Derek's first recording sample**

For his approved sample below, Derek setup a blanket tent and sat closer to the microphone.

**Derek's approved recording**

The difference between the first and second samples is easy to hear. The audio is no longer reverberating of the room, and Derek's sounds clear and up front. The audio is now ready to be mixed and edited. Below is the final result.

**The final audio track**

For the final audio, we've taken the raw audio recording, edited out the mistakes and breath sounds and mixed it to sound louder and clearer. So you can see how getting a good audio recording from the start can make the final product that much better!

What about AI voices?

VoiceMap tours are built on authenticity. Listeners choose self-guided audio tours over printed guides or generic apps because they want a human connection, someone who knows a place intimately and can share it in their own words. That's why we ask you to record your own voice rather than hire a voice artist, and it's also why we don't use AI to narrate tours.

That said, AI audio does have a place on the platform, in specific circumstances where it genuinely serves the tour and the listener.

Character voices

If your tour features a character, a pirate, a historical figure, a local voice distinct from your own narration, an AI-generated voice can bring them to life in a way that's genuinely entertaining. You can do this yourself using ElevenLabs, and we'll have more guidance on that soon. The key distinction is that character voices are a creative addition, not a replacement for your narration. Depending on your publishing plan, your editor may also be able to help with this, so speak to them if you'd like to explore it.

If English isn't your first language

Recording in a second language is hard, and we know it. If you're struggling with your English-language recording, you have two options. If you have an ElevenLabs account, you can clone your own voice and generate the English audio from your script directly. Alternatively, speak to your editor. With your permission, we can handle the cloning and generation for you. Either way, tours produced this way are labelled "AI generated voiceover" so listeners know what they're hearing.

Translated tours

When an English-language tour is translated into another language, we can clone your voice, with your permission, and use it to generate the translated audio. This means the tour still sounds like you, just in a different language. If you have an ElevenLabs account, you can also do this yourself. Either way, speak to your editor if you'd like to explore it. Translated tours using cloned audio are labelled "AI generated voiceover" too.

Updating a published tour

If your tour needs a script update and you're not able to re-record, whether that's due to time, circumstance, or the original recording environment no longer being available, we can clone your voice and implement the changes so the revision blends seamlessly with your existing audio. Speak to your editor to arrange this.

A note on quality

Not all AI voices are equal. A poor one sounds flat, with word stresses in all the wrong places. That gap is the difference between an immersive experience and one that's simply unlistenable. When we use AI audio, we're careful to choose voices that mimic human intonation and cadence as closely as possible.

We've found that cloning a publisher's own voice tends to produce the most satisfying results. Something of the original human inflection carries through, which matters to us and to listeners.

If you source your own AI voice, we reserve the right to decide whether the quality meets our standard for publication. This isn't about being difficult. It's about making sure every listener has a good experience. For tours distributed privately, we can be more flexible.

Any tour where AI audio is used for the main narration, rather than for character voices or minor revisions, is labelled "AI generated voiceover" on the tour page.