Frequently asked questions
Who runs VoiceGrid?
VoiceGrid is an independent storefront for a cloud text-to-speech API. We help indie developers and chatbot creators integrate high-quality speech into their apps via a simple API, providing a reliable audio engine without infrastructure headaches.
Where do the voices come from?
Our eight voice presets are based on licensed recordings from LibriTTS-R (CC BY 4.0) and public-domain audiobook readings from LibriVox. These sources provide the foundation for our warm, clear, and energetic voice options.
Is the API official?
We are not affiliated with OpenAI, ElevenLabs, Google, Amazon, Microsoft, or any other brand mentioned in comparisons. Our API mimics the OpenAI speech endpoint structure, but we are an independent service.
How do I contact support?
Support is by email only. We do not offer a phone number or live chat. You can reach us via the contact page or by emailing our support team directly.
Do you store my data?
Input text is used only to produce the audio and is not used for training. Generated audio is returned in the response and not stored. Only request metadata is kept for billing.
Is there an SLA?
We do not provide an SLA, uptime guarantee, enterprise contract, or certifications like SOC 2, HIPAA, or GDPR certificates. Our service is designed for indie developers without strict enterprise requirements.
Can I use the voices commercially?
Yes, you can use the generated audio commercially. The voices are based on licensed recordings and public-domain sources.
How are voices licensed?
Voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and public-domain audiobook readings from LibriVox. Every generated file carries an inaudible watermark marking it as synthetic speech.
What is the data retention policy?
Input text is used only to produce the audio and is not used for training. Generated audio is returned in the response and not stored. Request metadata is kept for billing purposes.
Do you have a physical address?
We do not list a physical address. Contact us via email for any inquiries.
Are you affiliated with OpenAI?
No, we are not affiliated with OpenAI. Our API endpoint shape matches the OpenAI speech endpoint, so their SDKs work by changing the base_url, but we are a separate, independent service.
How do I get help?
You can get help by contacting us via email. We do not offer phone support or live chat. Visit our contact page or email our support team for assistance.
Can I request a new voice?
We currently offer eight presets based on licensed recordings. We do not offer custom voice cloning or uploading your own voice. New voices are added based on our internal roadmap.
What happens to my data?
Input text is used only to produce the audio and is not used for training. Generated audio is returned in the response and not stored. Request metadata is kept for billing.
Who we are
VoiceGrid is built by a small team of indie developers who understand the friction of integrating voice into apps. We designed this API specifically for chatbot creators and voice assistant builders who need a reliable, straightforward TTS engine without the complexity of enterprise infrastructure.
Independent team
We are not a large corporation. Our team works directly with the code, ensuring that every feature serves the developer’s workflow. We prioritize practical utility over marketing fluff, focusing on what matters for your integration.
Developer focused
Our API mimics the familiar OpenAI request shape, allowing you to use standard SDKs by simply changing the base_url. We provide clear documentation and ready-to-use curl examples so you can go from sign-up to synthesized audio.
- JSON body compatible with OpenAI SDKs
- Simple
POSTendpoint for speech synthesis - Direct access to voice presets and languages
No corporate overhead
Without layers of management or complex billing departments, we keep operations lean. This allows us to offer transparent pricing and email support, ensuring your voice assistant project remains cost-effective and easy to manage.
Voice licensing
VoiceGrid’s audio engine is built on licensed recordings from the LibriTTS-R dataset, ensuring high-quality synthetic voices without complex licensing overhead for developers. This foundation allows us to offer consistent, clear audio output that integrates seamlessly into your chatbots and voice assistants.
LibriTTS-R source
Our voices are derived from the LibriTTS-R dataset, a comprehensive collection of spoken English and multilingual text. This dataset provides the raw material for our eight distinct voice presets, from the warm narrator aria to the deep documentary tone of orion.
CC BY 4.0 licensed
The underlying audio data is licensed under Creative Commons Attribution 4.0 (CC BY 4.0). This means you can use the generated audio in your commercial and non-commercial projects.
Public domain readings
Many of the base recordings come from LibriVox public-domain audiobook readings. This heritage ensures that the core vocal timbres are free from restrictive copyright claims, allowing indie developers to scale their voice-enabled apps without legal ambiguity.

Data policy
VoiceGrid respects your data privacy by design. Input text is used strictly to produce the audio file and is never used for model training or improving our engine. Once the audio is generated and returned in the response, it is not stored on our servers.
No training on input
We do not consume your input text to train or fine-tune our speech models. Your text remains yours, used solely for the immediate synthesis task.
No audio storage
Generated audio files are returned directly in the API response and are not persisted on our storage. We do not keep copies of your spoken audio for later use or analysis.
Metadata for billing
We retain request metadata, including timestamp, character count, and status, strictly for billing accuracy and dispute resolution. This limited data ensures you are charged correctly for the 1 million character rate.
Technical limits
We operate without the bureaucratic weight of large enterprises, which means we don’t offer formal Service Level Agreements (SLAs) or guaranteed uptime percentages. Our infrastructure is designed for simplicity and cost-efficiency, prioritizing direct developer access over corporate compliance structures. This approach keeps our pricing fair and our operations agile.
No SLA guarantee
We do not provide legally binding SLAs defining specific penalty structures or compensation for downtime. Our service is provided "as is," meaning we strive for reliability but do not contractually guarantee specific performance metrics like 99.9% availability.
No uptime promise
While we aim for consistent service, there is no formal promise of continuous uptime. Brief interruptions may occur during maintenance or infrastructure adjustments, and we do not compensate for these periods as we do not maintain complex enterprise support tiers.
Best effort support
Support is by email only. We do not offer live chat, phone numbers, or dedicated account managers. Our team does its best to assist via email, responding to issues as capacity allows, which keeps our overhead low and our API affordable for indie developers.
- Email support only
- No live chat
- No phone number

Contact and support
We know developers need answers when integrations break. Our support team is dedicated to helping indie creators and chatbot builders resolve API issues, billing questions, and voice selection queries quickly.
Email only support
Send your questions to our support address. We prioritize clear, technical responses that help you move forward with your integration. Since we are a small, independent team, we focus on quality over quantity.
No live chat
We do not offer real-time live chat. This allows our engineers to focus on solving your specific technical problems rather than managing multiple concurrent chat sessions. You will receive a thoughtful reply via email.
No phone number
There is no phone support line. Email is the most effective way to share logs, code snippets, and context. This asynchronous approach ensures you get a complete answer without waiting on hold.

Why we built this
VoiceGrid exists to remove the friction of integrating high-quality text-to-speech into indie apps and chatbots. We wanted a solution that feels like a natural extension of the developer’s toolkit, not a complex enterprise platform.
By focusing on a simple API and transparent pricing, we let you spend time building features rather than managing voice infrastructure or negotiating hidden costs.
Simplify TTS integration
We designed the API to mirror the OpenAI endpoint structure, allowing you to use existing SDKs with just a base_url change. This means you can drop voice capabilities into your workflow without learning a new paradigm.
Send text, receive audio. No complex pipelines, no streaming WebSocket dependencies, no custom SDKs to install.
- Single endpoint for all voices
- Standard JSON request format
- Works with existing OpenAI SDKs
Fair pricing model
We charge $0.01 per 1,000 characters, ensuring costs scale linearly with your usage. There are no monthly subscriptions, no minimum spend, and no surprise fees for failed requests.
This model lets indie developers and small teams predict their exact costs without overpaying for unused capacity or enterprise-grade SLAs.
Developer first approach
Every decision, from our TRC-20 USDT top-ups to our 4096-character input limit, is made with the developer’s workflow in mind. We prioritize clarity and control over corporate bloat.
We built this because we needed a reliable, affordable engine for our own projects, and we’re sharing it with the community that helped us get here.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Design conversational experiences that feel natural for users interacting with voice assistants and chatbots daily.
23 languages
Any voice speaks any of these languages.
