Sign inGet started
Dataset · Hugging Face

laion/Humaneness-Voice-Large-Alpha-Continuation-Listening

Humaneness Voice Large: additional 20/40/40 RL listening outputs 70 exact matched scripted takes from the separately identified continuation experiment. The released 1.4 Alpha adapter was continued for 2,000 additional successful updates, retaining AdamW moments with a new warm-up/cosine schedule: 20% Content Enjoyment, 40% matching two fixed supported source-audio emotion heads and 40% genuineness. Sixteen candidates per group, two reference/no-reference cohorts, 32 GPUs. This… See the full description on the dataset page: https://huggingface.co/datasets/laion/Humaneness-Voice-Large-Alpha-Con

Author / organizationlaion
License (as declared)cc-by-4.0
Downloads0 (as reported by Hugging Face)
Likes0
Created24 October 2026
Last updated at source24 October 2026
Embed this entry
<iframe src="https://silytics.si/embed/item/14c92a60-6b99-4d7d-96ec-6cd587c301e6" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

More from laion

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 24 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.