How to Use AI Voices in CapCut: Text-to-Speech and Voice Effects

To use AI voices in CapCut, add a text layer with your script, select it on the timeline, tap or click Text-to-speech, pick a voice, and apply it; CapCut then generates a voiceover clip on your timeline. If you want to change a voice you have already recorded instead, select the audio clip and use CapCut’s voice effects (voice changer) presets.

Below are the detailed steps for each version of CapCut, plus the fixes CapCut itself recommends when voices fail to generate or disappear from the list. Menu labels in this guide reflect what CapCut and several independent guides describe as of November 2026; CapCut updates its interface often, so expect small differences between versions, platforms and regions.

Two kinds of “AI voice” in CapCut

People searching for this usually want one of two things, and CapCut handles them with different tools:

Text-to-speech creates new audio. Voice effects modify audio that already exists. You can also combine them: generate a narrator with text-to-speech, then add an effect to that clip.

How to add a text-to-speech AI voice on mobile

These steps follow the flow described by CapCut’s Help Center and independent guides from EaseUS and Murf.

1. Open CapCut, tap New project, and import your video.
2. Tap Text in the bottom toolbar, then Add text, and type your script. Keep each text clip to a sentence or two so you can time it easily.
3. Tap the text layer on the timeline to select it. This step matters: CapCut’s Help Center states that text-to-speech only works on text layers, and nothing is generated if no text is selected.
4. In the bottom menu, tap Text-to-speech. You may need to scroll the toolbar sideways to find it.
5. Browse the voice list and tap a voice to preview it.
6. Tap the checkmark to apply it to that clip, or use Apply to all to convert every text clip in the project into speech with the same voice. EaseUS places the Apply to all toggle in the bottom-left corner of the voice panel.
7. A new audio clip appears under your video. Drag it to adjust timing, and delete or hide the on-screen text if you only wanted the voice.

A note on character limits

Sources disagree here. At least one guide says CapCut typically caps text-to-speech at around 1,000 characters per text clip, while others simply advise watching for a counter or warning in the panel. Splitting long scripts into several shorter text clips avoids the issue either way and gives you more control over pacing.

How to add a text-to-speech AI voice on desktop

The desktop editor (Windows and Mac) works the same way, but the controls live in a side panel.

1. Import your footage and drag it onto the timeline.
2. Open the Text tab, add a default text clip, and type your script.
3. Click the text clip on the timeline to select it.
4. In the properties panel on the right, open the Text to speech tab. Guides describe it sitting alongside tabs such as Basic, Animation and others; the exact position has moved between versions.
5. Choose a voice and click Start reading. CapCut processes the text and places a matching audio clip on the timeline.
6. Select the generated audio clip to adjust volume, speed and other settings in the audio panel.

The browser version of CapCut follows a very similar pattern: select a text layer, then use the Text-to-speech option in the editing panel.

How to change a voice with voice effects

If you recorded narration yourself and want it to sound different, use voice effects rather than text-to-speech. CapCut’s own voice-changing guide describes the mobile flow roughly as follows:

1. Start a new project and choose the video (or audio) whose voice you want to change.
2. Select the clip on the timeline. If the voice is part of a video clip, you can extract or separate the audio first so the effect only touches the sound.
3. Scroll the bottom toolbar and open Audio, then Voice effects. Some CapCut pages call this Audio effects instead, so look for either label.
4. Tap presets to preview them, then confirm the one you want.

VoxBooster’s walkthrough lists presets such as Chipmunk, Robot, Echo, Monster and Deep Voice, and notes that some effects may need an internet connection the first time they download on mobile. On desktop, select the audio clip and look for the voice changer or voice effects section in the audio properties panel; we could not find an official source confirming its exact location in the current desktop build.

Tips for better-sounding AI voiceovers

• Write for the ear. Short sentences and natural punctuation help the voice pause in sensible places.
• Fix mispronunciations with spelling. If a brand name or unusual word comes out wrong, spell it phonetically in the text layer, then hide the text if needed.
• Generate per clip. One voice clip per sentence or scene is far easier to re-time than one long block.
• Check the volume mix. Lower background music under narration so the voice stays clear.
• Preview before applying to all. Apply to all is quick, but test a voice on one clip first.

Voices not working or missing? Try these fixes

CapCut’s Help Center has articles on both problems. For text-to-speech that will not generate, it lists these common causes:

• Network problems. Text-to-speech runs in the cloud, so an unstable connection can interrupt it.
• Unsupported language or voice. Some voice and language combinations may fail.
• No text layer selected. The feature only works on text layers.
• Outdated app or browser, or a temporary bug.
• Audio is muted or misplaced. The voice may have been generated but muted, or placed outside the visible timeline range.

For voices that are missing or unavailable, CapCut points to temporary server maintenance, gradual feature rollouts, or an outdated app or browser. Its suggested order of fixes:

1. Re-check the voice panel and refresh.
2. Update CapCut to the latest version from the official app store or CapCut’s website.
3. Clear the cache or app data.
4. Log out and log back in.
5. Try a different network.
6. If the problem continues, contact CapCut Support through the app with a screenshot.

Free vs Pro voices

CapCut’s official help pages we reviewed do not spell out which voices are free. Several third-party guides say the free version gives you a set of standard voices, while some higher-quality, more natural-sounding voices require a CapCut Pro subscription. Voice lists also vary by region and rollout, which is why two people can see different options. Look for lock or Pro labels in your own voice panel, and treat any specific voice count you see online as potentially out of date.

The bottom line

For a narrator, use Text-to-speech on a selected text layer. To transform an existing recording, use Voice effects on an audio clip. If the option is missing, update the app and make sure a text layer is selected; if a voice disappears, CapCut says it is often a rollout or maintenance issue rather than something you did wrong.