Made with Wavemaker on 2026-10-23 on staging: the chatterbox-voice-convert graph on one L40S in one job (23 billed GPU-seconds, 2.3 credits). The source is the English take of the Northside Coffee voiceover and the target a 7 s clip of another Chatterbox voice; both voices are synthetic. The output is 5.96 s from a 5.95 s source, its transcript is unchanged, and its median pitch moved from 102 Hz to 254 Hz (the target's is 232 Hz). Its loudness follows the quieter target, so it is normalised before mixing.

Wavemaker-authored example. The download holds the brief and the delivered media; model weights, graphs and tools are not included.
