Dictivo — Keep your flow. (35 seconds, 2026-09-11)

Human speech recordings
Yamagishi, Junichi; Veaux, Christophe; MacDonald, Kirsten. (2019).
CSTR VCTK Corpus: English Multi-speaker Corpus for CSTR Voice Cloning Toolkit
(version 0.92). University of Edinburgh, Centre for Speech Technology Research.
Copyright (c) 2019 Junichi Yamagishi.
https://doi.org/10.7488/ds/2645
Creative Commons Attribution 4.0 International (CC BY 4.0)
https://creativecommons.org/licenses/by/4.0/
Full license: VCTK-license.txt in this directory.

Recordings: p225_156_mic1.flac and p225_103_mic1.flac.
Original human recordings, not synthesized speech or cloned voices.
Changes: decoded to WAV, gain adjusted, gently compressed and mixed with music.
No change to the spoken words, human voice speed or pitch. No endorsement.

Music and brand narration are AI-generated, separate from the human recordings.
Music: Lyria 2. Brand narration: Eleven v3, requested voice Aria.
Motion graphics, 3D brand assets and editing: Dictivo marketing project.

The displayed transcripts were produced by Local and Cloud Fast processing of
these recordings. The film presents those results in a designed interface with
shortened waits; it is not a continuous screen recording or a speed/comparative
accuracy benchmark. Scene titles are illustrative. Local keeps dictation audio
on-device after setup; Cloud Fast uploads the selected audio for transcription.

Reuse with this attribution, the license link, change notice and no-endorsement
notice included in the accompanying description or page.