How to calibrate voice for your content

Who can use this feature

Supported on Team and Enterprise plans
Only org admins and team admins can calibrate voice profiles. Team members can use voice profiles to generate content.
Also available in Chrome, Edge, Mac, Windows, and Figma.

With voice calibration, you can create outputs that sound just like you, faster. Savvy brands know how important consistent brand and individual, shared voices are. Voice allows you to build your different voices in Writer, and see them in action across the platform in the content you generate.

How to create a voice profile

Navigate to Setup > Content Generation > Voice. Select Tune your default voice.

There are two ways to tune a voice in Writer:

Analyze example copy: Give Writer examples of your best, most representative content. Writer will analyze the copy to identify distinguishing traits, vocabulary, and style.
Manually select traits and options: Let Writer know which characteristics and traits should make up your voice profile by selecting from a variety of options.

Creating a voice by analyzing copy

Submit copy that best represents your brand voice. The copy doesn’t need to come from the same source – you could combine high-performing emails with blog posts and press releases, for example. You must submit at least 300 words to tune your voice, and you can submit up to 1000 words. When you’re finished, select Next.

Writer will analyze your voice based on the copy provided, with a short overview and a longer, detailed profile. If the results sound like a good interpretation of your brand voice, select Next. If you want Writer to try again, select Re-analyze until it sounds right.

Finally, provide a Voice name and instructions for When to use the voice. Remember, you can create multiple voices for different use cases and audiences. In the example above, we’re creating a voice for our technical users – chief information officers, data scientists, and AI enthusiasts who want to read a complex blog post about cutting-edge technology. We can create more voices for less technical audiences, like new end users.

Creating a voice by manually selecting traits

You can create a voice manually by identifying a mix of traits that best represent your brand voice, and selecting a few preferences.

Select between 3-5 traits that best represent your brand voice. You can mix and match between Casual vs Formal, Tone, and Personality, or use all 5 traits within a single category.
How would you like to refer to your organization, your audience, and yourself? Let us know whether you have a preference.

Once you’re finished, select a sample text and select Apply voice to this text.


Sample text with no voice applied.	Sample text with voice applied.

Switching between the two sample texts will allow you to assess whether the selected traits and preferences accurately generate content that sounds like you. Continue making adjustments to your selections until you’re satisfied with the results under With voice applied.

Finally, provide a Voice name and instructions about when to use this voice: what types of users? what types of content?

How to manage voice profiles

On the voice management page, you’ll see the type of voice profile it is (generated via analysis or manually constructed), along with the user who created the profile and the date they did so.

Toggle Show in rewrite to display this voice as an option when rewriting content. See more below.

Select the ••• menu to access the following options:

Edit voice details (Name and When to use)
Edit the voice calibration itself (updating the copy or analysis, or reselecting traits)
Make default voice (the first voice you create will be the default unless you change it)
Delete the voice completely

How to apply voices to out-of-the-box apps

When using any out-of-the-box app, you’ll have the option to select a voice from a dropdown menu. You also have the option to select the “Writer voice” which uses neutral business language.

You can even generate multiple drafts using different voices:

Error messages generated using our casual voice for new users.

Error messages generated using our formal voice for technical users.

How to rewrite content using a voice

You can use the rewrite function to apply your voice to any content. Select the content you wish to rewrite, and select the voice you wish to apply.

NOTE: You must toggle on Show in rewrite in the Voice setup page for a voice to show up under the rewrite menu.

The rewrite menu will display your content rewritten in the selected voice. You can select the Copy icon to copy it to your clipboard – useful when you want to compare two versions – or select the black checkmark to replace your original content.

Rewriting content using Writer extensions

Any Writer extension which offers rewrite functionality can rewrite text in a voice profile: