Original JARVIS narration text, synthesized with Kokoro v1.0 / af_heart.
Model and synthesis library: Apache-2.0, hexgrad/Kokoro-82M and hexgrad/kokoro.
Model card: https://huggingface.co/hexgrad/Kokoro-82M
Library: https://github.com/hexgrad/kokoro
Training-data acknowledgements retained from the model card:
Koniwa tnc (<1h): CC BY 3.0, https://github.com/koniwa/koniwa
SIWIS (<11h): CC BY 4.0, https://datashare.ed.ac.uk/handle/10283/2353
No custom voice clone, paid TTS endpoint or HeyGen output is included.
Word timing comes from the model's predicted phoneme durations, using Kokoro's
native join_timestamps method. It is model timing, not an independent ASR pass.
Model weights and voice embeddings are not redistributed in this lesson bundle.
