Creating Audio
The Audio feature turns your uploaded documents into professional, narrated audio content. The process has two steps: first, the platform generates a script from your documents; then you render that script into an audio file using your chosen voice or voices.
Audio Formats
Section titled “Audio Formats”Before starting, choose the format that matches your goal:
| Format | Style | Speakers | Approximate Length | Description |
|---|---|---|---|---|
| Brief | Narration | 1 | ~4 minutes | A concise single-narrator summary of the key points |
| Deep Dive | Dialogue | 2 | ~12 minutes | An in-depth exploration with two hosts discussing the key findings |
| Critique | Narration | 1 | ~5 minutes | Constructive feedback and analysis of the content |
| Debate | Dialogue | 2 | ~10 minutes | Two perspectives debating the themes in your documents |
Each format supports a length setting: Shorter, Default, or Longer.
Step 1: Generate a Script
Section titled “Step 1: Generate a Script”The first step produces a structured script based on your documents. No audio is generated yet — just the written content you can review and edit.
- Open the Studio and select your project.
- Click New Audio and choose your preferred format.
- Set your options:
- The script generation step remains in English.
- Length — Shorter, Default, or Longer.
- Steering Prompt (optional) — up to 500 characters to focus the output. For example: “Focus on the competitive analysis section only.” If you leave this blank, the platform will auto-detect the main topics.
- Title (optional) — a custom title for this artifact. Auto-generated if not set.
- Source File Filter (optional) — restrict generation to specific files in the project.
- Click Generate Script.
The platform will search your documents, find the most relevant content, and write a script. You can monitor progress in the Studio. When it completes, the script is ready to review.
Step 2: Review and Edit the Script
Section titled “Step 2: Review and Edit the Script”Before rendering, review the generated script for accuracy and tone. This is the right moment to make corrections — editing the script costs nothing and voice synthesis only begins in the next step.
See Script Editing for guidance on what to review.
Step 3: Render the Audio
Section titled “Step 3: Render the Audio”Once you’re happy with the script, render it into an audio file.
- In the script view, click Render Audio.
- Choose the Render Language for the finished audio.
- Choose your voice:
- Voice A — the primary voice, used for all narration and the first speaker in dialogue formats.
- Voice B — the second speaker in dialogue formats (Deep Dive and Debate only).
- Adjust the audio settings if needed:
- Format — WAV (highest quality), MP3, or other compressed formats.
- Speed — adjust how fast the voices speak.
- Volume — raise or lower the output level.
- Click Synthesise.
The platform synthesises each line of the script using the selected voices, assembles the full audio track, generates a cover image, and saves a grounding report alongside the file.
Language Notes
Section titled “Language Notes”- Script writing and script editing stay in English so review remains consistent.
- The Render Language selector only affects the final audio synthesis step.
- The selector exposes the full currently supported language list for the active text-to-speech workflow.
What You Get
Section titled “What You Get”When the render job completes, the artifact includes:
- The audio file in your chosen format.
- A transcript — the plain-text version of the script as spoken.
- A cover image — an AI-generated thumbnail for the audio piece.
- A grounding report — an audit trail of every factual claim and its source document.
Advanced Usage Notes
Section titled “Advanced Usage Notes”Use these patterns to get more predictable results:
- Keep inputs specific and scoped to one intent per generation request.
- Save intermediate outputs often, especially before running major edits or regenerations.
- Use review and assignment workflows early to reduce rework late in the lifecycle.
Troubleshooting
Section titled “Troubleshooting”If results are not what you expected:
- Verify source quality first (missing context, inconsistent structure, or conflicting inputs reduce output quality).
- Re-check selected options and audience/profile settings before re-running.
- If a job fails repeatedly, wait briefly and retry once; then escalate with the artifact or job context.
Frequently Asked Questions
Section titled “Frequently Asked Questions”Why did quality change between runs?
Section titled “Why did quality change between runs?”Different input context, settings, and model/runtime conditions can produce different outputs.
Should I regenerate or manually edit?
Section titled “Should I regenerate or manually edit?”Regenerate when the core structure is wrong; manually edit when only wording, ordering, or style needs adjustment.
What is the best way to reduce iteration cycles?
Section titled “What is the best way to reduce iteration cycles?”Start with clear scope, run smaller batches, and use review checkpoints after each major step.