AudioStack now enables expressive voice rendering by default, so every new voice clip uses our most natural, lifelike text-to-speech models from the moment you start creating — no settings changes needed.
What has changed
Previously, expressive voice was available but had to be enabled manually via a toggle. It is now on by default for all supported voices, giving you richer, more natural-sounding audio straight away.
How it works
Expressive voice uses our latest generation TTS models to produce more natural, dynamic speech. The toggle remains visible in the toolbar so you can turn it off at any time for a specific clip or project.
How to use it
Open your project in InstantCreative as normal — expressive voice is already active for supported voice clips.
To turn expressive voice off for a specific clip, click the expressive toggle in the toolbar for that clip.
To turn it off for your initial ad generation settings, use the toggle in the voice selection modal before generating.
Tips
If you are using a voice with a strong regional accent, you may notice a small reduction in accent precision when expressive mode is on. If accent accuracy is critical for your project, try toggling expressive off for those clips.
If you encounter an audio artefact such as a pop or click, use the re-render option on that clip to generate a clean take.
Known limitations
Regional accent preservation is currently not fully supported in expressive mode for some voices. This is being improved at the voice model level.