Jessica
English · Explainer
For instructional designers and L&D teams
Paste a branching scenario or a role-play script. FullCastBook finds every character, gives each a different AI voice, and exports one audio file per scene or branch for your authoring tool.

Lewis Carroll
Alice's Adventures in Wonderland
Alice asks the grinning Cheshire Cat for directions — and gets none.
The Cat only grinned when it saw Alice. It looked good-natured, she thought; still it had very long claws and a great many teeth, so she felt it ought to be treated with respect.
Step-by-step guide
No card to sign up, nothing to install.
Hear an example
Generated from the script below with the studio's own voices and settings, one voice per speaker, joined the way an export joins lines. No edits afterwards.
Narrator · Dana · Sam · Coach · Voices: Gemini 3.8 Flash TTS
eLearning Scenario Voice-Over
What you get, from manuscript to finished audiobook.
Script template
Replace the text in brackets with your own, paste it into the studio, and every speaker gets a voice.
Copy this, replace the placeholders in square brackets, and paste it in or save it as a DOCX. Notes like [frustrated] are delivery directions; keep them. Each heading becomes its own MP3.
# Scene 1: [SITUATION IN A FEW WORDS] NARRATOR: [Set the scene in one or two sentences: who, where, what's at stake.] [CUSTOMER]: [frustrated] [Opening line: the problem, in their words.] [EMPLOYEE]: [A neutral first response.] [CUSTOMER]: [A detail that raises the stakes.] NARRATOR: What should [EMPLOYEE] say next? # Branch 1A: [BEST CHOICE] [EMPLOYEE]: [calm] [Acknowledge the feeling, then the next step.] [CUSTOMER]: [calmer now] [Reaction that shows it worked.] COACH: [Why this works, in one or two sentences.] # Branch 1B: [PLAUSIBLE BUT WRONG] [EMPLOYEE]: [confident, too quick] [The tempting mistake: over-promising, skipping a check.] [CUSTOMER]: [Reaction that hints at the problem later.] COACH: [What went wrong and what to try instead.] # Branch 1C: [CLEARLY WRONG] [EMPLOYEE]: [flat, dismissive] [The response that breaks the rule or the relationship.] [CUSTOMER]: [sharper, interrupting] [Escalation.] COACH: [The consequence, and the policy it breaks.]
Keep each line under about 25 words. Put the exact policy wording on screen, not in the voice-over.
Use cases
Where authors and storytellers put it to work.

In practice


Demo performances. In the studio you cast from standard and premium voice lists.
English · Explainer
English · Creator
English · Creator
English · Character
English · Character
English · Narration
English · Narration
English · Narration
English · Narration
English · Narration
In depth
eLearning scenario voice-over is what turns a branching scenario from a reading exercise into practice. Learners hear the customer's irritation, the colleague's pressure or the manager's hesitation, and they respond to the tone as well as the words. That's hard to get from on-screen text, and expensive to get from a recording session with several actors every time a branch is added or a policy changes.
FullCastBook gives you a cast from a script. Paste the scenario, and every character is detected and given a different AI voice, with a narrator for any setup text. You review the cast, add direction where a line needs it, and export the audio scene by scene. It's built on the same studio as our multiple voices text to speech tool, organised here for the way training scenarios are actually built: lots of short scenes, branches that reuse the same characters, and frequent revisions.
Authoring tools attach audio slide by slide, so the useful unit is a scene or a branch, not one long track. Write each one under its own heading ("Scene 1: The call", "Branch 1A: Acknowledge first", "Branch 1B: Jump to the fix") and put a speaker name before every line. Each heading becomes a chapter, and the ZIP export gives you one MP3 per chapter, named and ordered the way your storyboard is.
There are no per-line files, so if a single slide needs just one line, give that line its own heading, or split the MP3 in an audio editor. The export is ordinary MP3, which most authoring tools, including Articulate Storyline, Adobe Captivate and Rise, let you import onto a slide or block. FullCastBook doesn't publish SCORM packages or connect to your LMS; it produces the voice tracks your course is built around.
Every scenario has the same three jobs: the other person (a customer, a colleague, a candidate), the employee whose choices the learner controls, and someone who frames or debriefs the scene, such as a narrator or a coach. Give them voices that differ clearly in pitch, age and energy, so a learner who glances away from the screen still knows who's talking. Keep the coach calm and even; it's the voice learners will trust for the right answer.
Think about who the learner is, too. If your audience is front-line staff, the employee character should sound like a peer, not a polished presenter. Audition two or three voices per role before you settle, since voice samples are free to hear. Once chosen, a character keeps that voice in every branch of the project, which matters more than you'd expect: a customer who changes voice between branches breaks the illusion that the learner's choice affected the same person.
The wrong branch has to sound wrong. If the employee over-promises, rushes or dismisses a concern, write it that way and direct it: [confident, too quick], [flat, reading a script], [defensive]. Give the customer a reaction that rises or falls with the choice: [calmer now] after a good response, [sharper, interrupting] after a poor one. A [sigh] in the right place does more than a paragraph of feedback text.
Keep lines short and spoken. Real people don't recite policy in full sentences, and learners can't hold a forty-word line in their heads while choosing an answer. Put the exact policy wording on screen and let the characters talk the way people talk. More on writing these notes is in AI voice emotion tags, and if you want to hear a full script read through before you build the slides, an AI table read works the same way.
Training content changes: a refund window moves from 14 to 30 days, a product gets renamed, a regulator updates guidance. Edit the affected line, regenerate only that line, and export the chapter again. Product names, acronyms and internal jargon are worth checking early; write them the way they should be said, and see the pronunciation guide for the techniques that work.
Be clear about the limits. FullCastBook makes audio only: no video, no avatars, no lip-sync, and no caption or timing files. Your script is already the transcript, so paste it into your authoring tool's closed captions. There's no voice cloning, so you can't reproduce a particular trainer's or executive's voice; you cast from the standard and premium voice lists.
Signing up is free and needs no card. Generating audio uses credits, counted by the characters voiced, so a short scenario costs little; regenerating a line uses credits again. Compare the plans against the size of your course library.
As between you and us, you own the audio you export and can use it under our terms of service, in internal training or in courses you build for clients. Make sure you have the rights to the script you voice, and don't write scenarios that imitate a real, identifiable person without their consent.
Tips
FAQ
Something not covered here? Email us at support@fullcastbook.com
Yes. Paste the scenario with a name before each line, and every character is detected and given a different AI voice. You check the cast, add direction where needed and export the audio.
Put each scene or branch under its own heading. Each heading becomes a chapter, and the ZIP export gives you one MP3 per chapter. There are no per-line files, so give a single line its own heading or split it in an audio editor.
Yes, as ordinary MP3 files that you import onto a slide or block. FullCastBook doesn't publish SCORM or connect to an LMS; it makes the voice tracks.
Yes. Write a note like [frustrated, talking fast] or [nervous, hesitant] before a line, and add sound tags such as [sigh] where they help. Regenerate the line if the delivery misses.
Edit the affected line, regenerate only that line and export the chapter again. The rest of the approved audio stays the same.
No. It makes audio only. Your script is the transcript, so paste it into your authoring tool's captions.
Signing up is free with no card, and voice samples are free to hear. Generating audio uses credits; a short scenario costs little. See pricing.
You own the audio you export, and use follows our terms of service, including client work. Make sure you have the rights to the script.
Paste a role-play, cast the voices and export one file per branch. Free to sign up; creating uses credits.