Frequently asked questions
Who runs this site?
SmartSay operates as an independent storefront for the same text-to-speech API. We are not affiliated with OpenAI, ElevenLabs, Google, Amazon, or Microsoft. Each site in our network functions independently, providing direct access to our TTS engine without third-party intermediaries or shared corporate branding.
Where do the voices come from?
Our eight voice presets are based on licensed recordings from LibriTTS-R, derived from LibriVox public-domain audiobook readings. These recordings form the foundation of our synthetic voices.
Is my data used for training?
No. Input text is used exclusively to produce the audio file and is not used for training our models. We respect your data privacy by ensuring that the content you send for synthesis remains isolated to that specific generation task without being repurposed for future model improvements.
Do you store audio files?
Generated audio is returned directly in the response and is not stored on our servers. We retain only request metadata, such as time, character count, and status, strictly for billing purposes. Your actual audio output is transient and does not persist in our long-term storage systems.
How can I contact support?
Support is available via email only. We do not offer phone support, live chat, or on-premise installation assistance. For technical issues or billing inquiries, please send an email to our support address. We aim to respond promptly to all customer inquiries submitted through this channel.
Is there a phone number?
No, there is no phone number for SmartSay. We provide support exclusively through email to maintain efficiency and scalability for our API customers. You can reach our team via the contact page on this site for any questions regarding your account or technical integration.
Are you affiliated with OpenAI?
We are not affiliated with OpenAI. Our API endpoint structure mimics the OpenAI speech endpoint for convenience, allowing you to use official OpenAI SDKs by simply changing the base URL. However, we are an independent provider with our own infrastructure, pricing, and voice models.
What is the license for the voices?
The underlying voice recordings are based on LibriTTS-R, which is licensed under CC BY 4.0. This public-domain foundation allows us to provide high-quality synthetic voices. Each generated file also contains an inaudible watermark to identify it as synthetic speech, ensuring transparency in audio usage.
Do you offer refunds?
Failed requests are free; if synthesis fails, the charge is returned in full. We do not offer manual refunds for successful generations. Your balance does not expire, and you can top up with USDT on the TRON network. No subscriptions or minimum spends are required.
How do I find my API key?
Your API key is displayed on the /get-api-key/ page after you sign in with Google or email. You can issue a new key at any time, which immediately invalidates the old one. The page also shows your balance, usage stats, and a ready-to-use cURL example.
What is the editorial team?
SmartSay is an independent storefront, not a media publication. We do not have a traditional editorial team. The content on this site is curated to provide clear, factual information about our TTS API for smart home and IoT developers, focusing on technical accuracy and integration details.
Do you have a privacy policy?
We retain request metadata (time, characters, status) for billing purposes. Input text is used only for audio generation and is not stored. Generated audio is not stored. We do not collect excessive personal data beyond what is necessary for account creation and service delivery via email or Google.
Is the API open source?
The API itself is proprietary. We do not offer an official SDK of our own, though the OpenAI SDK works by changing the base URL. The underlying voice models are not open source, but they are built upon the CC BY 4.0 licensed LibriTTS-R dataset.
Can I use the voices commercially?
Yes, you can use the generated audio commercially. The base recordings are from LibriVox public-domain readings, and we add value through synthesis. Each file includes a synthetic watermark. There are no restrictions on commercial use, and you pay only for the characters you actually generate.
Our mission for IoT startups
SmartSay exists to remove the friction between text and voice in connected environments. We engineer a deterministic API that integrates seamlessly into smart home hubs and IoT dashboards, ensuring your devices respond with clear, reliable audio without the overhead of complex cloud dependencies.
Our focus is on simplicity and speed. By providing a straightforward REST endpoint, we enable startups to deliver voice responses that feel native to their hardware, supporting everything from kitchen assistants to industrial automation systems.
Focus on simplicity
We strip away unnecessary complexity. Our API accepts text and returns audio files via a standard POST request, mirroring familiar patterns so your team can integrate quickly without learning new paradigms.
The system handles character limits and formatting internally, allowing you to concentrate on the user experience rather than the infrastructure.
- Standard JSON request body
- Short phrases take a few seconds to generate
- No complex configuration required
Support for connected devices
Designed for constrained environments, our API delivers audio files that are ready to play immediately upon receipt. This is critical for voice-enabled devices where the delay before playback can disrupt the user experience.
Whether your device is a wall-mounted panel or a portable sensor, the lightweight response format ensures smooth playback even on modest hardware.
Developer-first approach
We provide complete documentation and a ready-to-use curl example from the moment you sign up. Your developers can test the integration in minutes using the OpenAI SDK or plain HTTP.
Rate limits and pricing are transparent, with no hidden tiers or unexpected charges, allowing you to scale your IoT fleet with confidence.
Voice licensing and sources
SmartSay’s voice engine is built on the LibriTTS-R dataset, providing eight distinct presets for smart home audio responses. These voices are derived from licensed recordings of narrators reading classic literature, ensuring a consistent sound profile for your IoT devices.
LibriTTS-R dataset
The core of our synthesis engine relies on the LibriTTS-R dataset, which offers high-quality, pre-processed speech data. This foundation allows us to deliver clear, deterministic audio outputs without the variability of proprietary black-box models, ensuring your devices sound professional and consistent across all languages.
CC BY 4.0 license
Voice samples are distributed under the Creative Commons Attribution 4.0 International license. This allows you to use the generated audio in commercial products. We handle the attribution requirements, letting you focus on integrating voice features into your smart home dashboards and automation hubs.
Public domain sources
The underlying readings originate from LibriVox, a project of public domain audiobook readings. By leveraging these established recordings, we ensure that our voice presets are grounded in widely accepted, high-fidelity source material.
Data privacy policy
SmartSay respects your data integrity by using input text strictly for immediate audio synthesis. We do not retain your raw text for model training or future development, ensuring your content remains yours.
No text training
Your input text is processed solely to generate the requested audio file. It is not used for training.
No audio storage
Generated audio files are returned directly in the response and are not stored on our servers after delivery. This ensures that your synthesized voice data does not linger in our infrastructure.
Billing metadata only
We retain only request metadata, including timestamp, character count, and status, for accurate billing purposes. This minimal data is used exclusively to calculate charges based on the $0.01 per 1,000 characters rate.
Technical limitations
SmartSay prioritizes deterministic, low-latency responses for IoT edge devices by processing requests sequentially. This design ensures predictable billing and simple integration, avoiding the complexity of managing partial audio streams. While this approach may feel different from real-time streaming services, it guarantees that every request is accounted for and charged only upon successful synthesis.
Understanding these constraints helps architects build robust voice-enabled dashboards and automation hubs that rely on consistent, batched audio delivery rather than continuous media pipelines.
No streaming support
We do not stream audio chunks as they are generated. Requests are processed sequentially, meaning you receive the complete audio file only after synthesis finishes. This ensures predictable latency for short phrases, though longer texts may require splitting and joining files manually.
No SSML
Our API accepts plain text input without Speech Synthesis Markup Language (SSML) tags. You cannot control pacing, pitch, or pronunciation breaks via markup. Instead, rely on the eight preset voices and the speed parameter (0.25–4.0) to adjust delivery characteristics.
No custom cloning
We do not offer custom voice cloning or upload your own voice recordings. The eight presets—aria, nova, luna, sage, felix, milo, atlas, and orion—are based on licensed LibriTTS-R recordings. This ensures consistent quality across all devices without the variability of user-uploaded samples.

Support and contact
SmartSay provides reliable support for IoT developers who need straightforward assistance without navigating complex enterprise hierarchies. Our team is dedicated to helping startups integrate our TTS API into their smart home ecosystems efficiently, ensuring that your voice-enabled devices respond accurately and reliably.
Email support only
We handle all inquiries via email. This ensures that every question receives a thoughtful, written response that you can reference later.
No phone number
We do not maintain a phone line. Voice calls can be disruptive and difficult to schedule, especially for distributed engineering teams. Email allows us to provide detailed technical answers, including code snippets or configuration examples, directly in the conversation.
No live chat
Unlike many consumer-focused services, we do not offer real-time live chat widgets. This reduces overhead and keeps our focus on engineering. You can expect a consistent, documented response rather than a fleeting chat session that disappears when you close the browser.
How we operate
SmartSay operates as an independent storefront, meaning we are a distinct entity providing the same underlying TTS API to a network of sites. Each storefront manages its own accounts and keys, ensuring that our operational structure remains simple and transparent for IoT developers who need reliable, deterministic audio responses without corporate overhead.
Independent storefront
We maintain direct control over the API infrastructure, ensuring that request routing and key management remain straightforward for startups integrating voice into smart home hubs and dashboards.
No affiliation with major brands
We are not affiliated with OpenAI, Google, or Amazon. While our endpoint shape mirrors OpenAI’s speech API for easy SDK compatibility, we are an independent provider of licensed synthetic speech.
Transparent pricing
Pricing is strictly usage-based at $0.01 per 1,000 characters, with no hidden fees or monthly plans. We accept USDT (TRC-20) for top-ups, ensuring clear, predictable costs for scaling IoT applications.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence.
23 languages
Any voice speaks any of these languages.
Sources and further reading
- Wikipedia
- openai-python on GitHub
- LibriTTS-R paper (arXiv 2305.18802)
- W3C WAI: making audio and video accessible
- FFmpeg documentation
- Creative Commons Attribution 4.0
- MDN: Web Speech API
- Node.js documentation
- curl documentation
- TRON developer hub: TRC-20
- OpenAI API reference: create speech
- openai-node on GitHub
