Frequently asked questions
What is the character limit per request?
Each request supports up to 4096 characters of input text. For longer texts, split into parts of up to 4096 characters and join the audio files. Spaces and punctuation count toward this character limit, so plan your payload accordingly to ensure smooth synthesis without exceeding the maximum threshold per API call.
Can I use this API for voice assistants?
Yes, the API is designed for deterministic, low-latency responses suitable for voice-enabled devices and automation hubs. You can integrate it into dashboards and IoT systems by sending JSON bodies via POST requests. The system processes requests sequentially, ensuring consistent audio generation for interactive voice assistant workflows.
Does the API support streaming audio?
No, the API does not support streaming, WebSocket connections, or real-time partial audio. It returns the complete audio file in the response after synthesis is finished. You must wait for the full file delivery before playing the audio in your smart home device or application.
What languages are supported?
We support 23 languages, including English, Spanish, German, French, Chinese, Japanese, Arabic, Hindi, and Russian. Any voice can speak any of these languages. If you do not specify a language, the system guesses it from the script, defaulting to English for Latin text.
Is the free trial really free?
Yes, signing in with Google provides $0.50 of credit, which equals about 50,000 characters. This trial is valid for 7 days. There are no hidden fees or subscription requirements to access this initial testing amount.
How long does the trial last?
The free trial credit is valid for 7 days from the time you sign in with Google. After this period, the credit expires if not used. To continue generating audio, you will need to top up your balance using USDT on the TRON network.
What happens if a request fails?
If synthesis fails, the charge is returned in full. You are not billed for failed requests. This ensures that your balance reflects only successful audio generations. You can retry the request or adjust parameters like speed or voice without losing any credit due to errors.
Do you offer custom voice cloning?
No, we do not offer custom voice cloning for customers. You cannot upload your own voice or design custom voices. The API uses 8 presets based on licensed recordings from LibriTTS-R and LibriVox. Each generated file carries an inaudible watermark marking it as synthetic speech.
Can I use the same key for multiple devices?
Yes, you can use the same API key for multiple devices. Each account has one active key that can be issued to various IoT endpoints. If you issue a new key, the old one stops immediately, so manage your key rotation carefully across your device fleet.
Is there a rate limit for IoT devices?
Yes, there is a limit of 60 requests per minute per key. Before the first top-up, this limit is 10 requests per minute. Requests are processed sequentially, so under load, a request may wait in a queue. Plan your device concurrency accordingly to avoid delays.
What audio formats are available?
You can choose from mp3, wav, opus, flac, and pcm formats. These options allow you to select the best balance between quality and file size for your specific IoT hardware. Specify the desired format in the response_format field of your JSON request body.
Do I need an SDK to use the API?
No, we do not have an official SDK. The API follows the same request shape as the OpenAI speech endpoint, so you can use the official OpenAI SDKs by changing the base_url. Alternatively, you can use plain HTTP requests to send JSON bodies and receive audio files.
Is my text data used for training?
No, your 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. We keep request metadata like time, characters, and status for billing purposes, but your actual content remains private.
How do I top up my balance?
Top up your balance using USDT on the TRON network (TRC-20) only. The minimum top-up amount is $5. Credits are automatically added within minutes after the transfer. We do not accept cards, PayPal, or subscriptions, and there are no monthly plans or invoices with VAT.
First request
curl
curl https://api.playae77.com/v1/audio/speech \
-H "Authorization: Bearer $TTS_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"tts-1","voice":"aria","input":"Hello from the API.","response_format":"mp3"}' \
--output hello.mp3Why IoT startups choose our API
Smart home and IoT startups require audio responses that feel immediate and reliable. Our API delivers deterministic latency, processing text at roughly real-time speed—about 25 characters per second for long inputs. Short phrases typically take a few seconds to generate, ensuring your devices respond without perceptible lag during user interactions.
Deterministic latency
Generation speed scales with input length. While long texts process at approximately 25 characters per second of audio, shorter commands may take a few seconds. Requests are processed sequentially, so under heavy load, a request might wait in a queue, but the system remains stable and predictable for your device's workflow.
Simple HTTP integration
Integrate voice capabilities into your dashboard or automation hub with a standard POST request to our JSON endpoint. You send text, receive an audio file. The architecture is designed to fit seamlessly into existing IoT pipelines without requiring specialized infrastructure or complex middleware to handle the audio synthesis.
No complex SDKs
Skip proprietary libraries. Our endpoint matches the standard request shape, allowing you to use official OpenAI SDKs by simply changing the base URL. Plain HTTP requests work just as well, giving your engineering team flexibility to choose the tools that best fit their specific smart home environment.
Supported languages for global devices
SmartSay supports 23 languages, covering major global markets including Arabic, Hindi, Chinese, and English. This range allows IoT devices deployed in diverse regions to deliver localized audio responses without requiring separate language packs or complex backend logic.
The API handles language detection automatically when the language field is omitted, defaulting to English for Latin-script text. For non-English Latin text, explicitly setting the ISO code ensures accurate pronunciation and avoids misinterpretation of regional dialects.
23 languages including Arabic and Hindi
Our engine delivers clear, natural speech across 23 distinct languages. This includes high-demand markets like Hindi and Arabic, as well as European languages such as German, French, Spanish, and others. Every voice preset can speak any of these languages, ensuring consistent audio quality regardless of the target region.
- Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish
Automatic language detection
For simplicity, the API guesses the language from the text script if you do not specify one. Latin script text defaults to English. This reduces configuration overhead for devices that primarily serve English-speaking users or where the language is known contextually.
ISO code support
For precise control, pass the ISO language code in the request. This is essential for non-English Latin-script text to ensure the correct voice model is applied. It prevents the API from defaulting to English when processing phrases that look like Latin but are spoken in another language.
Voice presets for different use cases
SmartSay provides eight distinct voice presets, each optimized for specific interaction contexts in smart home and IoT environments. These voices are derived from licensed recordings, ensuring consistent quality without the need for custom voice engineering.
Neutral news voice for alerts
Atlas delivers a clear, neutral tone ideal for system notifications and status updates. Its straightforward delivery ensures critical information is understood quickly, making it suitable for alarm systems or routine device confirmations.
Warm narrator for stories
Aria offers a warm, clear narration style. The tone is engaging yet relaxed, enhancing the listening experience for long-form content.
Energetic promo voice
Felix brings an energetic, promotional tone to device announcements. Use this voice for marketing messages, feature highlights, or interactive prompts where a lively, upbeat presence is required to capture user attention.
Audio formats for embedded systems
Selecting the right audio container is critical for embedded devices with limited bandwidth and processing power. Our API allows you to choose the exact format your hardware can decode, ensuring seamless playback across diverse IoT ecosystems without unnecessary conversion overhead.
MP3 for standard playback
MP3 remains the universal standard for smart speakers and voice assistants. It offers an optimal balance between file size and quality, making it ideal for devices with constrained storage or network bandwidth.
WAV for raw processing
For devices requiring uncompressed audio for further signal processing or analysis, WAV provides lossless quality. This format ensures high fidelity for raw audio pipelines where compression artifacts might interfere with downstream tasks.
Opus and FLAC support
Opus and FLAC are available as response formats. Both formats give you flexibility to optimize for either bandwidth savings or perfect audio reconstruction.
How the free trial works
New accounts created via Google sign-in receive an automatic $0.50 credit, equivalent to approximately 50,000 characters of synthesis. This allows your IoT startup to validate voice quality and latency on real devices before committing to a payment method.
$0.50 credit on sign up
Upon signing in with Google, your account is instantly credited with $0.50. This balance covers roughly 50,000 characters of text-to-speech generation, providing enough volume to test multiple voice presets and audio formats across your smart home devices.
7-day validity period
The trial credit remains active for seven days from the moment of account creation. This window is sufficient for engineering teams to complete integration tests, verify deterministic latency, and confirm that audio responses meet your device specifications.
No credit card required
You can generate your first API key and start synthesizing audio immediately without linking a payment method. Note that accounts created using only an email and password do not receive this initial trial credit, so choose Google sign-in to access the free tier.

Pricing for high-volume devices
SmartSay’s pricing model is designed for scalability, charging $0.01 per 1,000 characters. This flat rate applies regardless of voice selection or audio format, making cost prediction straightforward for devices generating frequent status updates. Failed requests incur no charge, ensuring you pay only for usable audio.
There are no monthly minimums or subscription tiers. Whether your device generates 100 responses or 10 million, the per-character rate remains consistent. This structure eliminates waste, allowing IoT startups to scale audio capabilities directly alongside hardware deployment without fixed cloud overhead.
$0.01 per 1,000 characters
The standard rate is $0.01 per 1,000 characters, including spaces and punctuation. This predictable unit cost allows precise budgeting for high-volume audio generation in smart home ecosystems. Longer texts: split into parts of up to 4096 characters and join the audio files.
- Rate: $0.01 per 1,000 characters
- Includes spaces and punctuation
- Consistent across all voices and formats
Pay only for successful requests
If synthesis fails, the charge is returned in full. You are billed only when audio is successfully generated and delivered. This ensures that transient network issues or invalid inputs do not inflate your operational costs unexpectedly.
No monthly minimums
There are no monthly plans, seats, or minimum spend requirements. Balance does not expire, and unused credit remains available for future use. This flexibility supports startups with variable usage patterns, from prototype testing to mass deployment.
Technical specifications
The API exposes a standard POST /v1/audio/speech endpoint, accepting JSON bodies identical to the OpenAI speech format. By pointing the base_url to https://api.playae77.com/v1, existing OpenAI SDKs integrate seamlessly with your IoT stack using your unique tts-... bearer key. The official OpenAI SDKs work by changing base_url to https://api.playae77.com/v1.
OpenAI-compatible endpoint
Send a JSON payload containing model, input, voice, response_format, and optional speed or language parameters. The endpoint returns the synthesized audio file directly in the response body, ready for immediate playback or further processing by your device firmware.
- Endpoint:
POST https://api.playae77.com/v1/audio/speech - Auth:
Authorization: Bearer tts-... - Formats: mp3, wav, opus, flac, pcm
60 requests per minute
Each API key is rate-limited to 60 requests per minute, ensuring predictable performance for concurrent device interactions. New accounts start with a lower limit of 10 requests per minute until the first top-up is processed. This limit applies per key, allowing you to distribute load across multiple keys if your deployment requires higher throughput.
4096 character limit
Each request accepts a maximum of 4096 characters for the input field, including spaces and punctuation. For longer responses, split the text into chunks, generate audio for each part, and concatenate the resulting files on your device. This approach maintains deterministic latency while supporting extended announcements or storytelling sequences.
Data privacy for customer devices
When your smart speaker processes a command, the raw audio data is transient. Our API returns the synthesized speech directly in the response and does not store the generated audio file on our servers after delivery. This ensures that the audio output remains strictly tied to the immediate request lifecycle, minimizing the risk of long-term audio retention in your device’s data chain.
No training on input text
The text you send to generate speech is used exclusively for that specific synthesis task. We do not use your input data to train or improve our models. Your voice command text or narrative content remains yours, ensuring that proprietary instructions or personal details entered into your IoT devices are not repurposed for algorithmic development.
No audio storage
Generated audio files are returned in the HTTP response and are not persisted in our storage systems. Once the audio is delivered to your device or application, no copy remains on our infrastructure. This eliminates the need for complex data deletion protocols or long-term archival considerations for your audio assets.
Metadata only for billing
We retain only the necessary metadata to ensure accurate billing and service monitoring. This includes the timestamp, character count, and request status for each API call. No audio content or full text logs are stored beyond what is required to reconcile your usage and process payments.
Getting started quickly
Start by signing up at /get-api-key/ with Google or email. Your unique API key appears instantly, ready for immediate use without waiting for approval or manual verification.
With your key in hand, you can begin testing the endpoint in minutes. The dashboard displays your balance, usage metrics, and a ready-to-copy cURL example for rapid validation of your first audio generation.
The API follows the standard OpenAI-compatible structure, allowing you to swap the base URL and use familiar SDKs. This consistency reduces integration friction and lets your development team focus on device logic rather than learning a new protocol.
Generate API key in seconds
Visit the /get-api-key/ page to create your account via Google or email. Your API key is issued instantly, with no manual approval steps required.
The key is immediately active, allowing you to start sending requests to the API endpoint without delay. You can regenerate the key at any time if needed.
- Sign up with Google or email
- Key issued instantly
- No manual approval needed
Use standard curl commands
Use the provided cURL example to send your first request. The JSON body requires the model, input text, voice, and format.
Authentication is handled via the Bearer token in the Authorization header. Failed requests incur no charge, ensuring safe testing.
The API returns the audio file directly in the response, allowing easy piping to your device's audio buffer.
- Use the ready cURL example
- Send JSON body with auth
- Receive audio file directly
Integrate in hours
Integrate the API into your smart home hub or IoT device within hours. The simple HTTP interface requires no complex SDKs or dependencies.
Split long texts into 4096-character chunks if needed, and join the resulting audio files seamlessly. The deterministic latency ensures reliable timing for voice responses.
Test with the free trial credit of $0.50, valid for 7 days, to verify integration before committing to higher volumes.
- Simple HTTP interface
- No complex SDKs
- Fast integration timeline
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.
