Full-cast audio

Full-cast audiobooks: one voice per character

Most text-to-speech gives your whole novel a single voice. VoiceYourNovel attributes every line of dialogue to the character who speaks it, matches a distinct voice to each one, and keeps a steady narrator underneath.

Start your audiobook See pricing

Character detection, voice casting, and chapter rendering are all in the studio. Export chapter MP3s as a ZIP, or one M4B with chapter markers for audiobook apps.

How it works

Three steps from manuscript to full-cast audio

Upload the manuscript
TXT, EPUB, DOCX, or PDF. The text is split into chapters, and narration is separated from dialogue before any casting happens.
Dialogue becomes a cast list
Speakers are identified line by line — including aliases and unnamed references — and merged into one profile per character, with age and role cues attached.
Cast, direct, and preview
Each character gets an identity-matched voice suggestion. Audition alternates, assign your picks, and preview full chapters in the browser.

What makes it full-cast

The hard part of a multi-voice audiobook is not the voices — it is knowing who is speaking. VoiceYourNovel attributes dialogue line by line, following speech tags, alternating exchanges, and aliases, so "the Warden", "he", and the character's actual name all resolve to the same voice.

Every speaking character in the manuscript is surfaced into the cast list. You can merge duplicates, rename entries, and decide whether minor characters get their own voice or share one — the cast mirrors your book, not a fixed template.

The narrator stays grounded

In a good full-cast production, the narration is the floor the performances stand on. Prose between the lines is read by one consistent, measured narrator, so the character voices land as performances instead of noise.

Character voices are matched to age, role, and temperament rather than pushed into caricature — a weary captain sounds weary, not cartoonish — and every assignment persists across chapters and sessions.

Frequently asked questions

How is this different from ordinary text-to-speech?: Ordinary TTS reads the whole book in one voice. Here, dialogue is attributed to individual characters and each one is performed by a distinct, identity-matched voice, with a separate narrator for the prose.
How does it know who is speaking?: Speaker attribution follows speech tags, conversational turn-taking, and aliases across the manuscript. Where a line is ambiguous, you can review and correct the attribution before rendering.
How many characters can a book have?: The cast list mirrors your manuscript — every detected speaking character appears in it. Minor characters can share a voice or be merged if you prefer a tighter cast.
Can I change a voice mid-project?: Yes. Reassign the character's voice and re-render the affected chapters. Assignments persist between sessions, so the rest of the cast stays exactly as you set it.
Can I export the finished audiobook?: Cast detection, voice assignment, and chapter rendering all happen in the studio. Finished chapters export as audio files in a ZIP.

Give every character their own voice

Upload your manuscript and hear the difference a real cast makes — attributed dialogue, matched voices, and a narrator who holds the story together.