KABURI-TTS (model weights)
Copyright 2026 Dialogue WG, Research and Development Center for Large Language Models (LLMC), National Institute of Informatics

Licensed under the Apache License, Version 2.0 (the "License"); you may not use
these files except in compliance with the License. You may obtain a copy of the
License in the bundled LICENSE file or at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed
under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR
CONDITIONS OF ANY KIND, either express or implied.

This repository distributes model weights that derive from, or redistribute,
third-party material:

  acoustic/
    Derived from Irodori-TTS-500M-v2 (https://huggingface.co/Aratako/Irodori-TTS-500M-v2),
    MIT License, Copyright (c) 2026 Aratako. The MIT terms apply to this
    derivative as well:

      Permission is hereby granted, free of charge, to any person obtaining a
      copy of this software and associated documentation files (the "Software"),
      to deal in the Software without restriction, including without limitation
      the rights to use, copy, modify, merge, publish, distribute, sublicense,
      and/or sell copies of the Software, and to permit persons to whom the
      Software is furnished to do so, subject to the following conditions:

      The above copyright notice and this permission notice shall be included in
      all copies or substantial portions of the Software.

      THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
      IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
      FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
      AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
      LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
      FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
      DEALINGS IN THE SOFTWARE.

    The upstream model card also states ethical restrictions (no impersonation,
    no misinformation) that accompany the MIT license; users are asked to observe
    them. See the Irodori-TTS-500M-v2 model card.

  textconverter/
    Fine-tuned from llm-jp/llm-jp-3-440m (https://huggingface.co/llm-jp/llm-jp-3-440m),
    Apache License 2.0. The tokenizer files (tokenizer.json, tokenizer_config.json,
    special_tokens_map.json) are redistributed from that model unmodified.

  raster/, predictor/
    Trained by us on LLM-jp-Zoom1; no third-party material is redistributed here.

Not redistributed in this repository, but required at synthesis time:

  Semantic-DACVAE-Japanese-32dim (https://huggingface.co/Aratako/Semantic-DACVAE-Japanese-32dim),
  MIT License, downloaded at run time. The DACVAE code installed as a dependency
  (https://github.com/facebookresearch/dacvae) is licensed under Apache License 2.0.

  The acoustic model and the timing predictor load the tokenizer of
  llm-jp/llm-jp-3-150m (Apache License 2.0) at run time.

No speaker reference audio is included here. The demo audio published in the
GitHub repository (https://github.com/llm-jp/kaburi-tts) is NOT covered by the
Apache License 2.0 and is subject to the terms of the JVS corpus; see
https://github.com/llm-jp/kaburi-tts/blob/main/docs/LICENSE-audio.md