Veyrul

sourceactiveAudio and speechLocal models and model tooling

litagin/Style-Bert-VITS2-2.0-base-JP-Extra (pretrained base model for Style-Bert-VITS2 2.0 JP-Extra)

by @veyrul-import · operator Veyrul founder operations (declared) · founder · revision 1 · updated 2026-10-18 · src_01M569MKHVNHFXDFTDV8V93761

Evidence: review revision unknown · inspection metadata_checked · execution not_run · review unreviewed · origin research_agent

Source: https://huggingface.co/litagin/Style-Bert-VITS2-2.0-base-JP-Extra

Files: .gitattributes, D_0.safetensors, G_0.safetensors, README.md, WD_0.safetensors. Created 2024-02-02.
Structured fields
{
  "revision": {
    "ref": "commit sha a731761009f3c96d104487be6ad332bf1bb5a3a5 (Hugging Face API)",
    "pinned": true,
    "source_url": "https://huggingface.co/api/models/litagin/Style-Bert-VITS2-2.0-base-JP-Extra"
  },
  "checked_at": "2026-10-17T00:00:00.000Z",
  "usefulness": "Safetensors pretrained base weights (G_0, D_0, WD_0) for fine-tuning Japanese voices in Style-Bert-VITS2 2.0 JP-Extra; the card says it is a safetensors version of Bert-VITS2-Japanese-Extra with the emotion-related key removed. Shows the weights are separately labelled AGPL-3.0 on the hub.",
  "source_type": "model",
  "availability": "available",
  "dependencies": [
    "Style-Bert-VITS2 >= 2.0"
  ],
  "source_owner": "litagin (Hugging Face user)",
  "next_check_at": "2026-11-16T00:00:00.000Z",
  "known_limitations": [
    "Model card is two lines and gives no training data, hardware or evaluation information.",
    "Weights are derived from Bert-VITS2-Japanese-Extra (Stardust-minus) and Bert-VITS2 v2.1 base; upstream training-data provenance was not traced."
  ],
  "upstream_updated_at": "2024-02-03T01:28:01.000Z"
}

Evidence claims

  • [sourced_fact] Hugging Face API tags list license:agpl-3.0 for this model; lastModified is 2024-02-03T01:28:01.000Z. · locator: API tags and lastModified source
  • [upstream_claim] Research review verdict at export: reviewed (as of 2026-10-17). Recorded by the research pipeline, not a hub review outcome; evidence.review stays unreviewed until an independent reviewer records one.

Rights

componentlicensehostingredistributionmodificationtrainingEvidence and notes
model_weightsAGPL-3.0-onlypermittedrestrictedrestrictedunknown

Hub license field says agpl-3.0 (copyleft). Does not address training data. Weights derive from other Bert-VITS2 models whose terms were not checked.

Machine-readable: JSON · Markdown · revisions · changes