top of page

TALQu Model Terms of Use

This agreement stipulates the terms and conditions for the use of the "Model for TALQu" (hereinafter referred to as "this Model") produced by the VirVox Project (hereinafter referred to as "VirVox") and the "TALQu Output Audio" (hereinafter referred to as "Output Audio") created using this Model by the user (hereinafter referred to as "you").

By downloading this Model, you are deemed to have agreed to all the terms of this agreement and are legally bound by them.

If the whole or part of this Model is modified, or if an extension of this Model is made, or if a new set of terms and conditions has been established, the new terms and conditions shall take precedence.

Note that this agreement does not stipulate the usage conditions of the "Character," defined in the following section. The use of the "Character" is governed by the "VirVox Project Secondary Creation Guidelines."

Definitions

"Character" refers to the abstract concept represented by its name and the artistic works of painting by the VirVox Project.

"This Model" refers to the model for TALQu.

"Vocoder model" refers to the model for synthesis prepared for TALQu models.

"Voice Provider" refers to the voice actor who provided the voice for the "Character."

"User" refers to the person who utilizes the Output Audio.

Constraints on machine learning models and other deliverables

This model by Kurono Takehiro is distributed using the Haruqa (https://haruqa.github.io/) model as a base model with prior permission from Haruqa.
This model was created using the Haruqa corpus with prior permission from Haruqa.
It is prohibited to generate new models based on this model by transition learning, fine tuning, etc.

Kurono Takehiro's vocoder model (HiFiGANModel) was generated using Haruqa corpus data with prior permission from Haruqa.
It is prohibited to use the vocoder model as a pre-trained model or as a base model and to generate derived models from the vocoder model.

 


This model of Shirakami Kotarou was created based on FlatBaseModel ( https://github.com/Haruqa/tacotron2/releases ).

Amendments to the Agreement

The amended agreement becomes effective when VirVox announces it on its website, etc., and you continue to use or download the Voicebank after such announcement.

In principle, works released before the amendment of this agreement that would become violations due to the amendment are not retroactively affected.

However, in the case of particularly malicious content, VirVox may request the cessation of its publication.

Disclaimer of Warranty and Limitation of Liability

(1)VirVox does not guarantee that this model will be suitable for your specific purposes, have the expected functions, accuracy, usefulness, or be free of defects.

(2)VirVox shall not be liable for any direct or indirect damages arising from the use of the output audio of this model.

(3)VirVox shall not be liable for any loss of data, malfunction or damage to equipment, or any other damages you suffer in relation to the output audio.

(4)VirVox shall not be liable for any transactions, communications, or disputes between you and any third party related to the output audio.

License to Use

VirVox grants you the right to use this model, provided you comply with this agreement. If you do not agree, you may not use this model.

  1. If you are an individual or an organization without corporate status, you may use this model free of charge within the scope of non-commercial activities.

  2. If you are an individual or an organization without corporate status, and you use this model for non-profit and paid usage (recovering only the actual expenses), you are permitted to publish and distribute derivative works created using this model.

  3. If you are an individual or an organization without corporate status, you may earn revenue through video posting sites or similar platforms via advertising revenue, subscription services, or donations.

※However, if the model is used for promotion in a contract with a corporation, or as a means of advertising other content or products for profit, it will be considered a commercial activity and separate permission will be required.

Cases Requiring Usage Permission

If you wish to use the Output Audio for commercial purposes, or if you wish to use the Output Audio beyond the scope of the above license, you must obtain a separate license in advance.
In such a case, please contact one of the following contacts.

  1. Commercial use of the Model by an individual or unincorporated organization for commercial purposes beyond the scope of the above.

  2. When a corporation uses and utilizes this model for commercial or non-commercial purposes.

  3. When the Output Audio is used in a commercial facility.

  4. When a corporation uses and utilizes this model for educational purposes.

Prohibitions

The following actions are prohibited when using the output audio:

  1. Using the output audio in works or sites intended to promote crime.

  2. Publishing voice using this model that significantly damages the credit, quality, dignity, or image of the voice provider or "Character."

  3. Publishing theoutput audio which might violate the publicity, reputation or other personal rights of any third party including (but not limited to) any original voice provider contributing to this model.

  4. Using this model in content that significantly violates public order and morals.

  5. Falsifying the copyright information of this model and the "Character."

  6. Violating this agreement.

  7. Selling the output audio using this model in the form of audio materials or other formats.

  8. Creating and providing output audio at the request of a third party.

  9. Secondary distribution of the Model in whole or in part.

  10. Using this model for commercial purposes for dubbing other than "character". *Except in cases where the following "Non-application of Prohibited Items" applies.

  11. Using this model for purposes other than TALQu.

  12. Generating a new model based on this model by transition learning, fine tuning, etc.

  13. Use the audio output from this model for machine learning.

  14. Secondary distribution of the vocoder model.

VirVox or the voice provider may request the cessation of the publication of the output audio if they deem it inappropriate.

Non-application of prohibited items

The following actions do not fall under the Prohibited Items.

Using this model for dubbing other than "character" for commercial purposes with the permission of VirVox Project and the rights holder.

Governing Law

This agreement is governed by and construed in accordance with the laws of Japan.

If any provision of this agreement is in conflict with mandatory laws, only that conflicting part will be null and void and will be amended to the closest content within the scope of the mandatory laws.

Jurisdiction

Any disputes related to this agreement shall be exclusively governed by the Tokyo Summary Court or the Tokyo District Court as the court of first instance.

Contact

Please contact us from ​here.

Preamble
Definitions
Constraints of this model
Amendments to the Agreement
Disclaimer
License to Use
Cases Requiring Usage Permission
Prohibitions
Non-application of prohibited
Governing Law
Jurisdiction
Contact
bottom of page