Frequently asked questions
Who owns the voices?
The voices are based on licensed recordings from LibriTTS-R under the CC BY 4.0 license, sourced from LibriVox public-domain audiobook readings. This foundation provides the raw audio material used to generate the synthetic speech output delivered via our API.
Are the voices realistic?
The eight presets—such as Aria, Nova, and Atlas—are suitable for accessibility tools. Each generated audio file carries an inaudible watermark that clearly marks it as synthetic speech, ensuring transparency about the audio source.
Is the team based in the US?
The provided facts do not specify the geographic location of the development team or the company headquarters. As an independent storefront, we focus on delivering the API service without specifying our physical team location.
Do you have a physical office?
We do not maintain a physical office for customer visits. Support is provided via email only, and there is no phone number or live chat option available for direct communication with our team.
Can I request a new voice?
We do not offer custom voice cloning or the ability to upload your own voice. The API provides eight fixed presets, and we do not currently accept requests to design or add new voice options.
Do you hire voice actors?
The voice data originates from public-domain audiobook readings available through LibriVox. We utilize these existing licensed recordings rather than hiring specific voice actors for new recordings, keeping the process straightforward and cost-effective for developers.
Is the API open source?
The API is a cloud service; we do not provide an official SDK of our own. You can integrate using the official OpenAI SDKs by changing the base URL or using plain HTTP requests.
Do you offer enterprise plans?
We do not offer enterprise contracts, SLAs, or uptime guarantees. There are no monthly plans, seats, or minimum spends. You pay only for what you use, with no complex billing structures or invoices with VAT.
Can I get a refund?
Failed requests are free, with the charge returned in full if synthesis fails. For other cases, we do not specify a general refund policy beyond the return of funds for failed requests, as we do not offer subscriptions.
Who maintains the servers?
The text does not specify the exact infrastructure provider or server maintenance team. We operate as an independent storefront for the TTS API, managing the service directly without detailing the underlying hardware maintenance providers.
Is there a blog?
The provided facts do not mention a blog section on the site. The available pages are home, pricing, docs, get-api-key, about, and contact. We focus on providing the API documentation and direct support channels.
Do you use cookies?
The provided facts do not specify the cookie usage policy for the website. Input text is used only to produce audio and is not used for training, but specific website tracking details are not listed.
What is the editorial team?
We are an independent storefront. The site provides straightforward documentation and support via email.
Can I use my own text?
Yes, you can send any text up to 4096 characters per request. For longer texts, split them into parts and join the resulting audio files. Spaces and punctuation count toward the character limit.
Who we are
SayGov is built by an independent editorial team focused on practical accessibility. We design for NGOs and public services that need reliable synthetic speech without enterprise complexity.
Our team understands the specific constraints of public information systems, ensuring our API integrates smoothly into existing workflows for visually impaired and non-native speaker support.
Independent editorial team
We operate as a standalone entity, not a division of a larger tech conglomerate. This structure allows us to maintain direct control over pricing and feature development; we do not offer uptime guarantees.
Our small, dedicated team prioritizes stability and clarity over rapid, untested expansion. You deal directly with the people managing the infrastructure.
- No corporate bureaucracy
- Direct engineering oversight
- Transparent decision-making
Focus on public sector needs
We recognize that public services require predictable costs and clear data handling. Our API processes requests sequentially, making it suitable for generating accessibility audio for websites and apps.
We avoid enterprise lock-in, allowing NGOs to integrate our TTS into any platform without negotiating complex contracts or minimum spend requirements.
- Predictable per-character pricing
- No minimum spend
- NGO-friendly billing
Developer-first approach
Our API follows a familiar JSON structure, similar to OpenAI’s speech endpoint. Developers can use standard HTTP libraries or the OpenAI SDK by simply adjusting the base URL to https://api.b777x.info/v1.
We provide clear documentation and ready-to-use curl examples, reducing the time from sign-up to first audio generation.
- Standard JSON API
- OpenAI SDK compatible
- Clear documentation
Voice licensing and sources
Our voices are derived from the LibriTTS-R dataset, which utilizes recordings from LibriVox. These are public domain audiobook readings, ensuring a robust and diverse foundation for synthetic speech.
LibriTTS-R dataset usage
We use the LibriTTS-R dataset as our primary source. This provides high-quality audio samples that form the basis of our eight voice presets. The dataset provides the source material for our eight voice presets.
CC BY 4.0 compliance
The underlying recordings are licensed under Creative Commons Attribution 4.0. We respect these terms by attributing the source appropriately. This allows us to offer transparent, legally sound synthetic speech for public and NGO use.
LibriVox public domain recordings
LibriVox provides public domain audiobook readings. These recordings are free from complex royalty structures. This simplicity helps keep our pricing predictable for indie developers and public service projects.
Data privacy and usage
We process your text solely to generate the requested audio file. Once the synthesis completes, the temporary audio data is discarded and not stored on our servers. This ensures that your content remains ephemeral, reducing the risk of long-term exposure for sensitive public information or NGO records.
Input text not used for training
The text you send to the API is used strictly for immediate conversion into speech. We do not ingest your inputs into our training datasets, meaning your specific phrasing or data choices do not influence future model iterations or voice behaviors.
Audio not stored after generation
Generated audio files are returned directly in the response and are not persisted in our storage systems after delivery. This minimizes data retention risks for accessibility tools handling personal or proprietary public service content.
Metadata kept for billing
We retain request metadata, including timestamp, character count, and status, solely for accurate billing and usage tracking. This data is not used for product development or shared with third parties.
- Timestamp of request
- Character count processed
- Request status (success/failure)
Support and contact
We prioritize efficiency for developers who need straightforward answers without corporate overhead. Our support structure is designed to be simple and direct, ensuring you can resolve integration issues quickly.
We do not use live chat widgets or phone lines, which can create false expectations of instant availability. Instead, we rely on asynchronous communication that respects both your time and ours.
Email-only support
All inquiries are handled via email. This allows our team to provide detailed, written responses that you can reference later in your development workflow.
You can reach us at [email protected] for any questions regarding your API key, billing, or technical integration.
No live chat or phone
We do not offer phone support or real-time chat widgets on this site. This approach keeps our operational costs low and passes those savings on to you in the form of transparent pricing.
There are no dedicated account managers or sales representatives to navigate; you deal directly with the team maintaining the API.
Response times
We aim to respond to support emails as quickly as possible. During high-traffic periods, responses may take slightly longer, but we prioritize technical issues that block development.
For urgent billing or API key issues, please include your account ID in the subject line to help us locate your account quickly.

Technical limitations
The API returns complete audio files rather than streaming partial audio chunks. Each request is processed sequentially, meaning under load a request may wait in a queue before synthesis begins.
No streaming or WebSocket
Unlike real-time voice assistants, you receive the full audio file only after generation completes. This approach ensures consistent file integrity but requires buffering for playback applications.
- Requests are processed one after another.
- Queuing may occur during high load.
No SSML or custom lexicons
The API does not support Speech Synthesis Markup Language (SSML) or custom pronunciation dictionaries. You cannot specify pauses, emphasis, or phoneme overrides via the standard JSON body.
- Standard JSON input only.
- No SSML tags supported.
No voice cloning or design
We offer eight preset voices derived from licensed recordings but do not provide custom voice cloning or voice design features. Users cannot upload samples to create unique voice profiles.
- Eight preset voices only.
- No custom cloning available.
Our commitment to transparency
Transparency is the foundation of our API. We provide clear pricing and limits so you can predict costs accurately for your accessibility projects.
Our open API documentation ensures you understand every parameter and constraint before you integrate.
Clear pricing and limits
We charge $0.01 per 1,000 characters, with no monthly plans or seats. Each account starts with a $0.50 trial valid for 7 days. You can top up via USDT on the TRON network, with a minimum of $5.
Requests are limited to 60 per minute per key. Failed requests are free, ensuring you only pay for successful synthesis.
- $0.01 per 1,000 characters
- No monthly plans or seats
- 60 requests per minute limit
- Free failed requests
No hidden fees or SLAs
There are no hidden fees or subscription traps. Your balance does not expire. We do not offer SLAs or uptime guarantees, keeping the model simple and predictable for indie developers and NGOs.
Support is email-only. There is no live chat or phone line, reducing overhead and keeping costs low for your projects.
Open API documentation
Our API follows a standard shape, compatible with existing OpenAI SDKs by simply changing the base URL. You can find all endpoints, including voice and model lists, in our open documentation.
Every generated file carries an inaudible watermark marking it as synthetic speech, ensuring clarity for end-users in public information systems.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Allow users to choose their preferred voice or disable audio entirely.
23 languages
Any voice speaks any of these languages.
