USER GUIDE

From Script to Podcast

Everything you need to know about directing, recording, and producing your audio drama with AudioDrama.

// CONTENTS
→ Getting Started→ The Studio→ Voice Performance Tags→ Director's Takes→ Generate from Take (STS)→ Voice Design & Library→ Generating Audio→ Pacing & Gaps→ Keyboard Shortcuts→ Publishing

// GETTING STARTED

Sign in with Google to create your account. The onboarding flow guides you through three steps:

1
Import or Choose a Script
Upload a PDF, text file, or Fountain screenplay — or paste text directly. The AI parser extracts characters, scenes, and dialogue, injecting voice performance tags automatically. You can also start with a public domain template.
2
Cast Your Voices
Assign an AI voice to each character from the ElevenLabs or OpenAI voice libraries. You can also design custom voices from a text description. Casting is optional at this stage — you can skip to the studio and cast later.
3
Enter the Studio
You’re taken directly to the production studio where your script is laid out and ready for audio generation.
Voice direction in AudioDrama

// THE STUDIO

The studio is your main workspace. It shows your full script on the left with inline audio controls, and a voice panel on the right for managing character voices.

Each dialogue line shows:

  • The character name and dialogue text (double-click to edit)
  • A cyan clip bar showing generated TTS audio — click to play
  • A purple clip bar for recorded takes — click to play
  • A red record button (visible on hover) for per-line recording
  • A GENERATE FROM TAKE button when a take is selected

Sound effects appear inline as [SOUND: ...] lines with their own generation and playback controls.

// VOICE PERFORMANCE TAGS

AudioDrama uses ElevenLabs v3, which supports inline tags to control how lines are delivered. The script importer adds these automatically, and you can edit them by double-clicking any line. Click the ? button in the studio header for a quick reference.

Emotion: [happy] [sad] [angry] [whisper] [sarcastic] [crying] [frustrated]
Reactions: [laughs] [sighs] [chuckles] [clears throat]
Pacing: [pause] [long pause] [rushed] [stammers] [hesitates]
Punctuation: ... trailing off  |  CAPS emphasis  |  — interruption

Examples:

[whisper] Don’t move... it’s RIGHT behind you.
[laughs] You actually said that to her FACE?
[crying] I can’t believe he’s gone.
[pause] So what do we do now?
Recording director's takes

// DIRECTOR'S TAKES

As a director, you can record your own performance of any line to guide how the AI voice should deliver it. There are two ways to record:

1
Per-Line Recording
Hover over any dialogue line to reveal the red record button. Click to start recording, click again to stop. Your take appears as a purple clip bar below the line.
2
Scene Recording
Click REC in any scene header to record the entire scene in one pass. After recording, the AI automatically aligns your performance to the script lines using speech recognition, splitting your recording into individual takes.

You can record multiple takes per line. Click the T1, T2 badges to switch between takes. Hover over a take badge to reveal the × delete button.

// GENERATE FROM TAKE (STS)

This is the magic feature. Once you’ve recorded a director’s take, the GENERATE FROM TAKE button appears. It uses ElevenLabs’ Speech-to-Speech (STS) technology to:

  • Take your recorded performance as a reference
  • Apply the character’s assigned AI voice
  • Preserve your pacing, emotion, inflection, and delivery
  • Output a new clip that sounds like the character, performed your way

The result replaces the standard TTS clip and is labeled STS instead of TTS. This gives you precise directorial control over every line — the AI voice acts out your performance.

Note: STS requires an ElevenLabs voice. OpenAI voices don’t support speech-to-speech.

// VOICE DESIGN & LIBRARY

The Library page (top nav) has a voice browser and a voice designer.

1
Browse Voices
Filter by provider (ElevenLabs, OpenAI), search by name, preview any voice. See which characters are using each voice.
2
Design a Custom Voice
Click + DESIGN VOICE and describe the voice you want in plain English — e.g., “A warm female voice in her 30s with a slight British accent, confident and clear.” The AI generates several previews. Audition them, pick your favorite, give it a name, and save it to your library.
3
Cast in the Studio
In the studio, the voice panel on the right shows all characters. Click CAST next to any uncast character to open the full casting modal with search, filters, and preview.

// GENERATING AUDIO

The bottom bar has two generation buttons:

  • GENERATE VOCALS — generates TTS audio for all dialogue lines using each character’s assigned voice. Shows a progress bar, live stats (completed, cached, skipped), estimated cost, and runtime.
  • GENERATE SFX — generates sound effects for all [SOUND: ...] lines in the script.

Generated clips are cached — re-running generation only processes changed or new lines. If you edit a line’s text, its clip is automatically invalidated and will regenerate on the next run.

If no voices are cast, a prompt will guide you to cast at least one character before generating.

// PACING & GAPS

Between each dialogue line, hover to reveal a thin gap handle. Drag up or down to adjust the silence between lines (shown in milliseconds). The default gap is 400ms. These values are saved automatically and used during sequential playback.

Use the transport controls in the bottom bar (▶ play, ■ stop) to hear your entire episode played sequentially with your configured gaps.

// KEYBOARD SHORTCUTS

Play / PauseSpace
Edit line textDouble-click
Save editEnter
Cancel editEsc

// PUBLISHING

The publish workflow is under active development. Here’s what’s planned:
1
Mastering
Your assembled rough mix is sent to Auphonic for professional mastering — loudness normalization to podcast standard (–16 LUFS), noise reduction, EQ, and compression. Download the broadcast-ready file.
2
Transcript & Metadata
Review the auto-generated transcript, add episode metadata (title, description, tags), and write show notes with cast credits and content warnings.
3
Distribution
Publish directly to podcast platforms. Generate an RSS feed, submit to Apple Podcasts, Spotify, and other directories — all from within AudioDrama.

Ready to direct your first show?

Import a script and start producing in minutes.