Audio Node Narration
Create spoken narration from a written script with an AI voiceover generator workflow in an Audio node on the Astorie canvas. Enter your lines, select a supported speech voice or create one with Voice Design, and review pronunciation, pacing and emphasis alongside your video shots.
Lines that read smoothly on a page can sound rushed or flat when spoken aloud. Comparing raw article copy with a script written for an Audio node shows how sentence length and clear phrasing improve generated voiceover.
Start with a script broken into scene beats and an open project on the Astorie canvas.
Place an Audio node on the canvas and choose Text to Speech mode. If your project needs a custom synthetic voice character, you can also explore Voice Design from a written vocal description in the Audio node.
Paste or type your narration segment directly into the Audio node's text field. In the documented Audio node workflow, speech text is entered inside the Audio node itself rather than passed through a Text node connection.
Select an available speech model, such as the supported ElevenLabs options in your workspace, and pick a voice and any available delivery or language controls that fit your scene.
Run the Audio node and listen to the full clip from start to finish. Check word stress, pauses and duration against the corresponding video shot, then export approved audio segments for final mixing in your editor.
Check three details in every voiceover take before locking your edit timing.
Listen carefully to brand names, character names, technical terms and numbers. If a word sounds unnatural, adjust the spelling or punctuation in the Audio node and generate a fresh take for that line.
A voiceover that fills every frame leaves the viewer no time to register the image. Keep sentences concise so you can leave natural pauses around scene transitions and key visual reveals.
When you generate narration in scene-by-scene blocks, play consecutive clips back to back. Confirm that energy, speaking speed and vocal character feel consistent from the opening hook to the closing line.
Keep your script formatting, node settings and editorial mix aligned across the workflow.
| Input | A speakable narration script entered directly into the Audio node's text field, or a vocal description for Voice Design. | |
|---|---|---|
| Output | Generated speech audio clips on the canvas for review and export into your video editing timeline. | |
| Review | Check pronunciation of names and figures, natural sentence cadence, clip duration and tone consistency across takes. | |
| Plans and credits | Speech generation uses workspace credits based on the selected audio model and plan tier; review the pricing page for details. | |
| Rights and mixing | Final dialogue leveling, music ducking, loudness normalization and timeline sync happen in your video or audio editor. |
Connect spoken narration with documented audio models and video handoff tools on Astorie.
Review the documented modes and inputs for Text to Speech, Voice Design, Sound Effects, Music and Video to Audio.
See how Eleven v3, Multilingual v2, Turbo v2.5 and Dialogue v3 support narration and dialogue on the canvas.
Organize approved narration tracks and video takes for final assembly in your nonlinear editor.
In Text to Speech mode, the Audio node takes the script text you enter inside the node and synthesizes spoken audio using the model and voice you select. Because each Audio node sits on the same canvas as your image and video stages, you can keep narration takes organized right beside the shots they accompany.
Generating one scene or paragraph per Audio node makes timing and script revisions much easier to manage. If a single sentence changes during the edit, you only need to regenerate that short cue instead of re-rendering the entire voiceover track.
Available delivery controls depend on the speech model you select in the Audio node. For example, supported ElevenLabs models on Astorie offer different strengths for expressive delivery, dialogue or steady long-form narration; check the controls shown in your active node.
Spoken duration depends on word count and speaking cadence rather than a fixed video frame count. Review the generated audio clip against your picture and trim the script or adjust pause timing in your editing timeline to lock sync.
Open an Audio node, enter a short scene script and compare voice takes before assembling your final video cut.
