[Alpha] Text-To-Speech in the Template Editor

🪄 Alpha functionality: This functionality is still being developed and is expected to change over time. Right now we only enable it for a limited number of customers. If you're interested, reach out to your Customer Success Manager for more details and eligibility requirements.

---

You can turn any Text layer into a Speech audio layer in the Template Editor.

Speech layer options include several different narrators and Playback speed options to best showcase your product or service:

• Language acts as a filter for narrator options. When you select a language, you see only the narrators who match your selection.
• Narrator defines the narrator voice. You can either select a narrator from the list, or use a feed as input.
• Speed defines the speed of the narration. This can also be set through a feed.
• Volume controls the volume level of the speech layer.

Create a speech layer from a text layer

1. In the Template Editor, select a text layer.
2. In the panel on the right, click the Text-to-speech tab.
3. Set the speech options.
4. To preview your audio layer, click Play.
5. Click Create audio layer.
6. To edit general audio properties like start or end time, continue editing in the same panel. Your changes are saved automatically.

Edit a speech layer

You can edit an existing speech layer's general audio properties as well as the language, narrator and speed properties of the speech layer. These changes cause the layer to be re-generated and the  changes are applied immediately.

1. In the Template Editor, select the text layer.
2. Edit the audio and speech options.
3. To preview your audio layer, click the play button.

Create a speech layer from a feed

You can use a feed to generate variations with different narrators. When using a feed, there is no need to set the Language option. (Technically, selecting a language merely filters the list of available narrators.)

1. In the Template Editor, select a text layer.
2. In the panel on the right, click the Text-to-speech tab.
3. On the Narrator drop-down menu, select Content from feed column.
4. In the Feed column field, specify the source of the Narrator content.
5. On the Speed drop-down menu, select Content from feed column.
6. In the Feed column field, specify the source of the Speed content.
7. To preview your audio layer, click Play.
8. Click Create audio layer.
9. To edit general audio properties like start or end time, continue editing in the same panel. Your changes are saved automatically.

Tips for using Text-To-Speech

• Break up the script into snippets that you place in separate audio layers. This gives you more control over how the voiceover plays.
• Match the script length to the video length. The maximum speed is 1.5x, because beyond this, speech may become hard to follow. Don't try to squeeze 60 seconds of VO into a 30-second video.
• To create an accented VO, mix the narrator and the script language. For example, using a es-MX narrator to speak an en-US script results in accented English.
• Add periods (.) in the script to create pauses.
• To spell out the letters of a word, insert spaces: S A L E - SALE!
• The VO reads out symbols like /, @, % and #. If you put a number after # (#7), the VO reads it as "number" ("number seven").