Version: 2026-10 (Last updated: October 10, 2026)
1. Scope
These Model Terms are the per-model terms established by Jugemu Tech LLC (the "Company") under Article 4, Paragraph 4 of the Lisnify Terms of Service (the "Terms"). Users shall comply with the Model Terms applicable to the speech synthesis model (TTS engine) they configure for a show.
The Model Terms are established by the Company based on the license or terms of use set by each model's provider, and may be added to or changed independently of the Terms. Where the Model Terms and the Terms differ, the Model Terms prevail with respect to the use of the model concerned.
2. Models and their Model Terms
| Model (name shown in Lisnify) |
Provider |
Upstream license / terms |
Consent when configuring a show |
Model Terms |
| Gemini 2.5 Flash / Gemini 3.1 Flash (Preview) |
Google (Vertex AI) |
Google Cloud Terms, Generative AI Prohibited Use Policy |
Not required (link only) |
Gemini |
| OpenAI gpt-4o-mini-tts |
OpenAI |
OpenAI Terms of Use, Usage Policies |
Not required (link only) |
OpenAI gpt-4o-mini-tts |
| Supertonic 3 |
Supertone Inc. (model publisher) |
BigScience Open RAIL-M License |
Required |
Supertonic 3 |
| VOICEVOX |
The VOICEVOX project and the rights holders of each character |
VOICEVOX software terms, per-character voice terms |
Required |
VOICEVOX |
| Kokoro-82M |
hexgrad (model publisher) |
Apache License 2.0 |
Not required (link only) |
Kokoro-82M |
For models marked "Required", the provider's license or terms impose obligations on downstream users, so we ask you to agree to the Model Terms when you configure the model for a show. For models marked "Not required", we only show a link to the Model Terms and do not ask for confirmation, but you must still comply with them.
3. Provisions common to all models
- AI disclosure: The Company automatically appends a statement to the show description in each show's feed (the podcast RSS) that the script and audio of the show are generated by AI. Users must not remove this statement or present generated content as if it were produced by a human.
- Responsibility for content: Users are responsible for the content they generate (the RSS feeds they register, the show settings, and the use and publication of generated scripts and audio) as set out in Article 7, Paragraph 4 of the Terms. The Company does not review generated content in advance.
- Prohibited activities in the Terms: The prohibited uses in each Model Terms apply in addition to Article 8 (Prohibited Activities) of the Terms.
4. Procedure for changing the Model Terms
- When the Company adds to or changes the Model Terms, it will notify users who have configured the affected model for a show of the content of the change and its effective date at their registered email address.
- For changes that impose new restrictions on users, the Company will provide such notice at least 14 days before the effective date. Changes that relax restrictions may take effect upon notice.
- If a user continues to use the Service on or after the effective date, the user is deemed to have agreed to the change.
- If the Company becomes unable to provide a particular voice or model for reasons attributable to the provider (termination of a voice license, discontinuation of a model, etc.), the Company will stop offering the voice or model for new selection and notify the owners of shows that use it to change the voice or model before the next episode is generated. Until the change is made, episode generation for the show is recorded as failed (no credits are consumed). The Company will not automatically replace the voice or model a user selected with another one.
- The history of changes is recorded in the "Revision history" at the end of this index and of each Model Terms page.
5. Contact
For inquiries about the Model Terms, please use the contact information in Article 15 of the Terms.
Revision history
| Version |
Date |
Change |
| 2026-10 |
October 10, 2026 |
Initial version (Gemini, OpenAI gpt-4o-mini-tts, Supertonic 3, VOICEVOX, Kokoro-82M) |