NPC Audio Lab: TTS API for Indie Game Developers

By NPC Audio Lab editorial team · Updated 2026-11-19

NPC Audio Lab provides a dedicated text-to-speech API designed for indie developers and audio designers. This page outlines our editorial approach, the technical foundations of our voice models, and how we handle your data. We focus on delivering consistent character voices across 23 languages without the overhead of full voice casts.

If you are building games and need reliable, cost-effective speech synthesis, this is your resource. We explain our mission and our licensing model. Read on to understand how our API integrates into your workflow, what data we retain, and how to reach our support team via email.

NPC Audio Lab: TTS API for Indie Game Developers

Frequently asked questions

Who runs NPC Audio Lab?

NPC Audio Lab is an independent storefront for the 89999.info text-to-speech API. We are an editorial team focused on indie game developers and audio designers. We are not affiliated with OpenAI, ElevenLabs, or any other major brand. Our goal is to provide a simple, reliable API for generating character speech in 23 languages without the overhead of full voice casts.

Where do the voices come from?

Our eight voice presets—Aria, Nova, Luna, Sage, Felix, Milo, Atlas, and Orion—are based on licensed recordings from LibriTTS-R. This dataset uses public-domain audiobook readings from LibriVox. Each voice is a distinct preset designed for specific tones, such as warm narration or energetic promos, ensuring consistent character voices for your game projects.

Are the voices copyrighted?

The underlying recordings are sourced from LibriVox public-domain audiobook readings. However, every generated audio file carries an inaudible watermark marking it as synthetic speech. This watermark helps identify the audio as AI-generated without imposing complex copyright restrictions on the output files you receive for your game integration workflows.

Do I need to credit the voices?

The source material comes from LibriVox public-domain readings. While attribution is good practice, the primary identifier is the inaudible watermark in each generated file. You do not need to credit specific voice actors, as the voices are presets based on the LibriTTS-R dataset.

How is my data used?

Your input text is used solely to produce the audio response. It is not used for training our models or improving the API. We retain request metadata, such as time, character count, and status, strictly for billing purposes. Your text and generated audio are processed and returned without long-term storage for model improvement.

Is my text stored after generation?

No. Your input text is used only to produce the audio and is not stored after generation. We do not retain the actual text content for training or analysis. Only the metadata required for billing, such as the timestamp and character count, is kept. Your creative input remains private and transient.

Is my audio stored after generation?

No. Generated audio is returned in the response and is not stored on our servers. Once you receive the audio file, it is yours to use in your game. We do not keep copies of the synthesized speech files. This ensures your character lines remain your property without lingering data dependencies.

What data is kept for billing?

We keep request metadata for billing purposes. This includes the time of the request, the number of characters processed, and the status of the request. This data allows us to accurately calculate costs based on the $0.01 per 1,000 characters rate. No personal text or audio content is retained for billing.

How do I contact support?

Support is available via email only. There is no phone number or live chat option. You can reach us through the contact page on our site. We aim to provide clear, direct answers to your technical and billing questions. For urgent issues, email remains our primary and only channel for customer assistance.

Is there phone support?

No, there is no phone number for NPC Audio Lab. We do not offer phone support to keep our operations lean and focused on the API. All support inquiries are handled via email. This allows us to maintain detailed records of your issues and provide written solutions that you can reference later.

Is there live chat?

No, there is no live chat widget on our site. We rely on email support to ensure every query is addressed thoroughly. This approach avoids the pressure of real-time responses and allows our team to provide detailed answers to your questions about the API, voices, or billing without immediate time constraints.

Do you have a physical address?

We do not list a physical address. NPC Audio Lab operates as an independent digital storefront for the 89999.info API. Our team is distributed, and we conduct all business, including support and billing, online. You can find our contact email on the /contact/ page of our website.

Are you affiliated with OpenAI?

No, we are not affiliated with OpenAI. Our API endpoint mimics the OpenAI speech endpoint structure, allowing you to use OpenAI SDKs by changing the base URL. However, we are a separate entity offering our own text-to-speech engine. We do not use OpenAI’s models or infrastructure, and our pricing is independent.

Are you affiliated with ElevenLabs?

No, we are not affiliated with ElevenLabs. We are an independent storefront for the 89999.info API. While both services offer text-to-speech, our voice presets, pricing model, and infrastructure are entirely separate. We do not use ElevenLabs’ technology or branding, and our eight voice presets are based on LibriVox recordings.

Do you offer custom voice cloning?

No, we do not offer custom voice cloning for customers. You cannot upload your own voice or design a new voice preset. We provide eight fixed presets: Aria, Nova, Luna, Sage, Felix, Milo, Atlas, and Orion. If you need a specific voice, you must choose from these existing options for your game NPCs.

Do you have an SLA?

No, we do not offer an SLA or uptime guarantee. Our service is provided as-is without enterprise contracts. Requests are processed sequentially, so under load, a request may wait in a queue. We aim for consistent performance, but we do not guarantee specific availability windows or compensation for downtime.

Do you have certifications?

No, we do not have SOC 2, HIPAA, or GDPR certificates. We are an independent indie-focused API without enterprise compliance certifications. Our data practices are transparent: input text is not used for training, and audio is not stored. However, we do not hold formal third-party certifications for these practices.

How do I get my API key?

Sign up on the /get-api-key/ page using Google or email and password. Accounts are per site. Once registered, you receive a Bearer token (tts-...) for authentication. You can issue a new key at any time, which immediately invalidates the old one. Your key, balance, and usage stats are displayed on the dashboard.

Can I use the voices in a commercial game?

Yes, you can use the generated audio in commercial games. The voices are based on public-domain LibriVox readings, and the output is synthetic speech. You can use the generated audio in commercial games. You can integrate them into your game assets without additional licensing fees beyond the API usage.

What is the CC BY 4.0 license?

CC BY 4.0 is the Creative Commons Attribution license applied to the source recordings from LibriVox. It allows us to process and adapt the recordings for our TTS engine while maintaining transparency about our sources. Our voices are derived from this dataset, ensuring legal clarity for your game development projects.

Our Mission

Sound wave illustration for: Our Mission

NPC Audio Lab exists to bridge the gap between ambitious indie game narratives and the high cost of traditional voice acting. We provide a reliable, programmatic solution for generating consistent character speech, allowing developers to iterate on dialogue without the logistical overhead of studio bookings.

By leveraging a robust TTS API, we enable small teams to produce high-quality audio assets efficiently, ensuring that voice production remains accessible and scalable regardless of budget constraints.

Supporting Indie Developers

Supporting Indie Developers

We design our API specifically for the constraints of independent development. With a simple integration and transparent pricing, you can generate voice lines for dozens of characters without managing complex audio pipelines or negotiating with multiple voice actors.

Democratizing Voice Lines

Democratizing Voice Lines

High-quality voice acting is no longer reserved for well-funded studios. Our service allows solo developers and small teams to implement full voice casts in 23 languages, ensuring that narrative depth is determined by creativity rather than budget.

Simplifying Audio Production

Simplifying Audio Production

Instead of coordinating schedules and recording sessions, you send text and receive audio files. This streamlined workflow integrates directly into your game engine, letting you focus on gameplay and story while the API handles the vocal performance.

The Editorial Team

Sound wave illustration for: The Editorial Team

We are a small editorial team focused on the intersection of narrative design and technical audio. Our background in game audio production informs how we structure the API for indie developers who need consistent voice lines without the overhead of full voice casts.

Game Audio Enthusiasts

Game Audio Enthusiasts

We understand the specific needs of NPC dialogue, from looping ambient speech to distinct character reactions. Our team tests voice presets in actual game environments to ensure they integrate smoothly into typical workflows.

Technical Writers

We document the API clearly, focusing on practical integration steps for developers. Our guides prioritize concrete examples and clear endpoint structures, helping you generate audio files without unnecessary complexity.

Community Focused

We listen to feedback from indie studios to refine our voice presets and pricing. Our goal is to provide a tool that adapts to the real-world constraints of small teams and freelance projects.

Voice Licensing

Sound wave illustration for: Voice Licensing

Our voice models are built from licensed recordings sourced directly from LibriTTS-R, which draws from public-domain audiobook readings available on LibriVox. This foundation ensures that every synthetic voice carries a clear, traceable lineage without the complexity of multi-layered rights management.

We provide a straightforward licensing path for indie developers who need consistent character voices without navigating intricate legal hurdles for each asset.

LibriVox Source

All eight voice presets—Aria, Nova, Luna, Sage, Felix, Milo, Atlas, and Orion—are derived from high-quality recordings in the LibriVox archive. These sources offer diverse vocal qualities suitable for various NPC archetypes, from newsreaders to storytellers.

The underlying dataset, LibriTTS-R, aggregates these public-domain contributions, giving us a robust and legally clear foundation for synthetic voice generation.

  • Source: LibriVox public-domain audiobooks
  • Dataset: LibriTTS-R (CC BY 4.0)
  • Coverage: 23 languages supported

CC BY 4.0 License

The LibriTTS-R dataset is released under the Creative Commons Attribution 4.0 International license. This allows us to process and adapt the recordings for our TTS engine while maintaining transparency about our sources.

You can use the generated audio files commercially without additional attribution requirements beyond your standard game credits, though we encourage acknowledging the original LibriVox contributors.

Public Domain Readings

The source recordings are in the public domain, meaning their copyright has expired or been explicitly waived. This eliminates royalty disputes for the base vocal performances.

While our specific synthesis engine adds unique processing, the foundational vocal timbre remains rooted in these freely available historical readings, ensuring ethical and legal clarity for your projects.

Data Policy

Sound wave illustration for: Data Policy

Input text is used solely to generate the audio file and is not added to our training data. Generated audio is returned in the response and is not stored on our servers after delivery.

No Training Usage

Your character dialogue remains yours. We do not use input text to improve our models or train future voice generations.

No Audio Storage

Audio files are returned in the response and are not retained on our infrastructure after the response is sent. This keeps your game assets secure and private.

Metadata Retention

We keep request metadata, including timestamp, character count, and status, strictly for billing purposes. This ensures accurate charges and helps troubleshoot any generation issues.

Data Policy

Contact and Support

Sound wave illustration for: Contact and Support

NPC Audio Lab relies on email for all support needs, keeping communication simple and trackable for indie developers managing tight budgets and schedules. We avoid the noise of live chat widgets to focus on clear, documented resolutions for your API integration questions.

Our editorial team handles inquiries directly, ensuring technical accuracy without the filter of support tiers. You can reach us at [email protected] for billing issues, API errors, or voice selection advice.

Email Only

We do not offer phone support or live chat. All interactions happen via email, allowing you to attach logs, code snippets, or audio samples for precise troubleshooting.

This approach reduces overhead, keeping our costs low and our API pricing competitive for freelance developers and small game studios.

No Live Chat

We intentionally skip real-time chat widgets to maintain focus on detailed, written responses. This ensures you receive a clear record of the solution without waiting on hold during your development sprints.

Response Time

We aim to respond to all emails within 48 hours during weekdays. Complex technical queries may take longer as we verify API behavior against our rate limits and voice presets.

Why We Built This

Sound wave illustration for: Why We Built This

Traditional voice-over pipelines often stall indie projects due to budget constraints and scheduling bottlenecks. Securing consistent character voices requires coordinating multiple talent pools, which can delay production for weeks.

High Studio Costs

Hiring full voice casts for NPCs is expensive. Even for minor characters, studio time and per-hour rates add up quickly, making it hard for small teams to justify extensive dialogue trees.

  • Eliminates hourly studio fees
  • Reduces per-character casting costs

Scheduling Delays

Finding available talent and coordinating recording sessions creates significant lead times. Delays in voice recording can block level design and narrative implementation, pushing back release dates.

  • Instant generation vs. booking slots
  • No rescheduling needed

Scalability Issues

As dialogue grows, managing hundreds of voice files becomes unmanageable. Generating new lines or tweaking scripts requires re-recording, which is slow and costly compared to API-driven synthesis.

  • Generate unlimited variations
  • Easy script updates

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. This ensures consistent voice delivery across extended conversations without hitting request limits.

Avatar of the voice aria
ariafemale
Avatar of the voice nova
novafemale
Avatar of the voice luna
lunafemale
Avatar of the voice sage
sagefemale
Avatar of the voice felix
felixmale
Avatar of the voice milo
milomale
Avatar of the voice atlas
atlasmale
Avatar of the voice orion
orionmale

23 languages

Any voice speaks any of these languages.

Sources and further reading

  1. Wikipedia
  2. openai-python on GitHub
  3. Node.js documentation
  4. Creative Commons Attribution 4.0
  5. LibriVox public-domain audiobooks
  6. openai-node on GitHub
  7. IETF RFC 7845: Ogg Opus
  8. Tether (USDT)
  9. MDN: Web Speech API
  10. TRON developer hub: TRC-20
  11. LibriTTS-R paper (arXiv 2305.18802)
  12. W3C WAI: making audio and video accessible

More on this site