inklight-app/paced-voices

Inklight paced voices

Slower voice states for Kyutai's Pocket TTS, made for Inklight, an iPhone app that reads books aloud on the device. Each file is a voice state for the english_2026-04 model: the language model's key-value cache after hearing a speaker's reference clip. The clip was time-stretched before the model heard it, and the model imitates the clip's tempo, so a voice loaded from one of these files reads more slowly of its own accord, with the pauses a slow reader adds, and no time-stretching of the output.

Files

english_2026-04/<voice>@<pace>.safetensors, four paces for each of ten voices, 40 files, 369 MB in all, 7 to 12 MB each. The pace in the name is the speed the state was measured to produce relative to the voice at its own pace (0.9, 0.85, 0.8 and 0.75); every file was measured within about 0.02 of its name (the largest deviation is 0.020, most are within 0.015), over three sentences and three or six seeds, with the articulation rate taken from the rendered speech and pauses left out. stops.json lists, for every file, the clip stretch it was made from, the measured pace and its spread, and, where one was made, the word error rate of a check transcription.

The voice names are Kyutai's voice ids in Pocket TTS. The files are in the same format as Kyutai's own voice embeddings for that model (a per-layer cache and its offset) and are read the same way.

How they were made

The reference clip of each voice, the same recording Kyutai made the voice from (_ORIGINS_OF_PREDEFINED_VOICES in kyutai-labs/pocket-tts), was time-stretched with ffmpeg and encoded with the voice-cloning release of the model. How far a voice follows the stretch is the speaker's own, so for each voice the stretch was swept, each candidate state measured, and the one closest to each pace kept.

Credits and licence

These states are adaptations of recordings published under the Creative Commons Attribution 4.0 International licence, and are released under the same licence.

• anna, azelma, charles, eponine, fantine, george, michael, paul and vera are derived from the CSTR VCTK Corpus, version 0.92, by Junichi Yamagishi, Christophe Veaux and Kirsten MacDonald, The Centre for Speech Technology Research (CSTR), University of Edinburgh, doi.org/10.7488/ds/2645 (speakers p228, p303, p254, p262, p244, p315, p360, p259 and p229), licensed under CC BY 4.0, by way of the cleaned recordings in kyutai/tts-voices.
• alba is derived from the recording "Casual" by Alba MacKenna, kyutai/tts-voices, licensed under CC BY 4.0.
• The model that computed them is Pocket TTS by Kyutai, licensed under CC BY 4.0.

Changes made: each recording was slowed, without changing its pitch, to between 0.6 and 0.86 of its original speed, and reduced to a Pocket TTS voice state. Neither the speakers, CSTR, nor Kyutai endorse these files or the app. They are provided without warranty.