The voices behind the stories

These narrators are AI-generated. Preset names identify voice references; the source speakers did not personally record your story.

Pocket TTS

Speech model by Kyutai: Manu Orsini, Simon Rouard, Gabriel De Marmiesse, Václav Volhejn, Neil Zeghidour and Alexandre Défossez. English September 2026 preset model, licensed under CC BY 4.0. The inference software uses the MIT license. StoryPath uses the public preset-only model, generates new narration and converts it to MP3. Playback speed can be adjusted.

Model and license · Software · Voice provenance and licenses

Selected English presets

• Alba: Alba MacKenna, casual performance. Source · CC BY 4.0.
• Marius: Unmute voice donor Selfie. Source · CC0.
• Javert: Unmute voice donor Butter. Source · CC0.
• Peter: Peter Yearsley, Voice-Zero / LibriVox selection. Source · CC0.
• Fantine: CSTR VCTK Corpus, speaker p244. Source · CC BY 4.0.
• Eponine: CSTR VCTK Corpus, speaker p262. Source · CC BY 4.0.

VCTK credit: Junichi Yamagishi, Christophe Veaux and Kirsten MacDonald, CSTR VCTK Corpus. Enhanced references were supplied by Kyutai. Preset embeddings were computed by Kyutai. CC0 terms.

When distributing narration outside StoryPath, retain the applicable model and voice attribution. These credits do not imply endorsement by the speakers or model authors. Older recordings may use the previously installed Kokoro engine.