Frequently asked questions
How do I get my API key?
Sign up on the /get-api-key/ page with Google or an email and password. Your account page displays your active API key, balance, and usage details. Each site account maintains one active key at a time. You can issue a new key at any moment, and the old one stops immediately.
What is the character limit per request?
Each request supports up to 4096 characters of input text. Spaces and punctuation count toward this limit. For longer texts, split the content into chunks of 4096 characters or fewer and join the resulting audio files.
Can I change the speed of the voice?
Yes, use the speed parameter (0.25–4.0) to adjust playback. Slow down narration for clarity or speed up content for quick updates. Note that requests are processed sequentially, so under load, a request may wait in a queue before generation begins.
Which audio formats are best for assistive devices?
We support mp3, wav, opus, flac, and pcm. For assistive devices, MP3 or OPUS are efficient for bandwidth, while WAV offers uncompressed quality. Choose based on your device’s playback capabilities and storage constraints. Note that all generated audio carries an inaudible watermark marking it as synthetic speech.
How do I handle texts longer than 4096 characters?
Send each chunk of up to 4096 characters as a separate request. Join the returned audio files to create the final output. Note that requests are processed sequentially, so longer texts may take time. For best results, split text at natural pauses to ensure smooth audio playback when combined.
Is the API compatible with OpenAI SDKs?
Yes. The API uses the same request shape as the OpenAI speech endpoint, so official OpenAI SDKs work by setting base_url to https://api.playtk333.com/v1 and providing your AidVoice key. Note that we are not affiliated with OpenAI. This compatibility applies to standard text-to-speech requests; we do not offer our own SDK or streaming support.
What happens if I exceed the rate limit?
Your key allows 60 requests per minute, or 10 per minute before your first top-up. Requests are processed sequentially, so exceeding limits may cause queuing delays. If a request fails, it does not consume your character balance, ensuring you only pay for successful audio generation.
Do I need to install any specific SDKs?
No specific SDK is required. You can use the official OpenAI SDKs or plain HTTP requests. We do not provide our own dedicated SDK, so standard libraries work perfectly. Simply update your base URL to https://api.playtk333.com/v1 and use your API key. This approach works for web, mobile, or backend integrations without needing additional dependencies.
How do I set the language for non-English text?
Include the language parameter with the ISO code in your request body. While the API guesses the language from the script (defaulting to English for Latin text), explicitly setting this optional field ensures accurate pronunciation for non-English content, such as Spanish or Japanese, preventing misinterpretation by the system.
Can I use the same key for multiple projects?
Yes. Each account provides one active API key, which you can use across multiple applications, devices, or projects without needing separate credentials for each. Note that issuing a new key immediately invalidates the old one, so ensure your projects are updated before rotating keys.
How do I know if a request failed?
Failed requests are free, and the charge is returned in full. You can verify the status by checking the HTTP response code and body. We also keep request metadata, including time, characters, and status, for billing purposes. This record helps you track usage and troubleshoot any synthesis issues that may have occurred during processing.
Is there a webhook for completion status?
No, we do not offer webhooks. The API is synchronous, so you receive the audio file directly in the response. Since requests are processed sequentially, you may experience slight delays during high load, but you get the result immediately after the wait. Monitor usage and status via your account dashboard or standard API responses.
Can I stream the audio output?
No, streaming is not supported. The API returns the complete audio file after synthesis. For real-time needs, you may need to implement client-side buffering or pre-generate audio files for your content.
How do I reset my API key?
Generate a new key anytime via your account page. The new key activates immediately, and the old one stops working at once. This ensures secure access management for your projects without downtime. You can rotate keys as needed to maintain security for your assistive device or elderly care tools.
First request
curl
curl https://api.playtk333.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.mp3Python (openai)
from openai import OpenAI
client = OpenAI(base_url="https://api.playtk333.com/v1", api_key="tts-...")
with client.audio.speech.with_streaming_response.create(
model="tts-1", voice="nova", input="Hello from Python.", response_format="mp3"
) as response:
response.stream_to_file("hello.mp3")Node.js (openai)
import OpenAI from "openai";
import fs from "node:fs";
const client = new OpenAI({ baseURL: "https://api.playtk333.com/v1", apiKey: process.env.TTS_API_KEY });
const res = await client.audio.speech.create({ model: "tts-1", voice: "milo", input: "Hello from Node.js." });
fs.writeFileSync("hello.mp3", Buffer.from(await res.arrayBuffer()));| voice | |
|---|---|
aria | female |
nova | female |
luna | female |
sage | female |
felix | male |
milo | male |
atlas | male |
orion | male |
| language | |
|---|---|
ar | العربية |
zh | 中文 |
da | Dansk |
nl | Nederlands |
en | English |
fi | Suomi |
fr | Français |
de | Deutsch |
el | Ελληνικά |
he | עברית |
hi | हिन्दी |
it | Italiano |
ja | 日本語 |
ko | 한국어 |
ms | Bahasa Melayu |
no | Norsk |
pl | Polski |
pt | Português |
ru | Русский |
es | Español |
sw | Kiswahili |
sv | Svenska |
tr | Türkçe |
Authentication
AidVoice uses standard Bearer token authentication for all API requests. Include your unique API key in the Authorization header of every call. This key is generated when you sign up on our /get-api-key/ page and is required to access the speech endpoint.
Use Bearer token with your API key
Include your API key in the header as Authorization: Bearer tts-.... This key authenticates your account and tracks your usage against your balance. Ensure the key is kept secure, especially when embedding it in client-side code or CI/CD pipelines.
One active key per account
Each account maintains exactly one active API key at any time. When you generate a new key from your dashboard, the previous key is invalidated immediately. This ensures a single point of control for your assistive device deployments.
Replace key immediately if compromised
If your key is exposed, generate a new one from the dashboard. The old key stops working at once, ensuring your usage remains protected without interruption. Any characters consumed by the old key before invalidation are still billed to your account.
Speech Generation Endpoint
Send your text to the speech generation endpoint to receive synthesized audio. This POST request accepts a JSON payload containing the text, voice selection, and model preference. It is designed for straightforward integration into assistive devices and elderly care tools where reliable audio output is critical.
POST /v1/audio/speech
Direct your requests to the endpoint at https://api.playtk333.com/v1/audio/speech. This single URL handles all synthesis jobs. Ensure you use the correct base URL in your client configuration to avoid connection errors.
JSON body with model, input, voice
The request body must include the model ("tts-1" or "tts-1-hd"), the input text (up to 4096 characters), and the voice preset. You can also specify a language ISO code, response_format, and speed (0.25–4.0). If no language is set, the system guesses it from the script.
- model: "tts-1" or "tts-1-hd"
- input: your text string
- voice: e.g., "aria", "nova", "felix"
- speed: 0.25 to 4.0
Returns audio file directly
The API returns the generated audio file directly in the response body. There is no intermediate polling step; you receive the file immediately upon completion. This direct return simplifies playback integration in your application or device firmware.

Available Models and Voices
Our API offers two model tiers, tts-1 and tts-1-hd, which share the same underlying engine and pricing structure. This allows product teams building assistive devices to select audio fidelity based on their specific hardware constraints without complicating cost forecasts.
tts-1 and tts-1-hd models
Both models process text identically, delivering natural speech suitable for elderly care tools. The choice between them depends on your device’s processing power and storage limits rather than price differences.
8 preset voices for different tones
Select from eight licensed voices, each designed for specific use cases in assistive technology. Female options include aria (warm, clear narrator), nova (bright, upbeat), luna (soft, storytelling), and sage (calm, instructional). Male choices feature felix (energetic, promo), milo (friendly, conversational), atlas (neutral, news), and orion (deep, documentary).
Consistent pricing across models
Pricing is uniform across both models. You pay $10 per 1 million characters processed, regardless of which model you choose. This simplifies budgeting for your assistive device deployment.
Request Parameters
To generate speech, send your text in the input field. The API accepts up to 4096 characters per request, including spaces and punctuation. For longer texts, split your content into chunks of this size and join the resulting audio files later.
If the API guesses the wrong language, use the language parameter with a standard ISO code. Since Latin script defaults to English, you must specify the code for other Latin-based languages to ensure accurate pronunciation for your assistive device users.
Input text up to 4096 characters
Send the text to synthesize in the input field. The limit is 4096 characters per request, counting spaces and punctuation. If your content exceeds this, split it into multiple requests and concatenate the audio outputs.
This limit ensures consistent processing times. For long-form content, batching requests allows you to maintain control over the audio structure while staying within the API’s single-request constraints.
- Count all characters, including spaces and punctuation.
- Split long texts into 4096-character chunks.
- Concatenate output files for full playback.
Optional language ISO code
The API guesses the language from the script, defaulting to English for Latin text. If the guess is wrong, set the language parameter with the correct ISO code.
This is crucial for multilingual assistive tools. For example, Dutch or Portuguese text in Latin script will default to English without this parameter, leading to incorrect pronunciation.
- Defaults to English for Latin script.
- Set
languagefor non-English Latin text. - Supports 23 specific languages.
Speed control from 0.25 to 4.0
Adjust the speed parameter to control playback rate, ranging from 0.25 (slow) to 4.0 (fast). This helps tailor audio for elderly users who may need slower speech for clarity.
Fine-tuning speed ensures the audio is accessible and easy to understand, without altering the voice characteristics or pitch of the selected preset.
- Range: 0.25 to 4.0.
- Defaults to 1.0 (normal speed).
- Adjust for accessibility needs.
Response Formats
AidVoice returns the generated audio file directly in the response body, eliminating the need for external storage or complex download pipelines. You receive the complete audio data immediately, allowing your assistive device or elderly care application to play the speech without delay.
MP3, WAV, OPUS, FLAC, PCM support
The API supports five response formats: MP3, WAV, OPUS, FLAC, and PCM. MP3 and OPUS provide compressed audio suitable for general playback, while WAV and FLAC offer lossless quality. PCM provides raw, uncompressed linear audio data for systems requiring specific bit-depth handling.
Choose format based on device needs
Select the format that best matches your assistive device’s capabilities. For web-based interfaces or limited-bandwidth mobile apps, OPUS or MP3 reduces data usage. For local playback on hardware with robust codecs, WAV or FLAC ensures high fidelity. Use PCM if your audio hardware requires raw bitstreams.
- Use OPUS/MP3 for bandwidth efficiency
- Use WAV/FLAC for lossless playback
- Use PCM for raw hardware input
Audio returned in response body
The audio binary is included directly in the HTTP response body. There is no intermediate storage step, so your application receives the file instantly. This direct delivery ensures low latency for real-time speech generation in elderly care tools.

Rate Limits and Performance
AidVoice processes synthesis requests sequentially, meaning your audio generation happens one after another rather than in parallel. Under heavy load, a request may wait in the queue before processing begins, so your integration should account for potential delays rather than expecting instant returns. This behavior ensures stable throughput but requires your application to handle asynchronous or queued responses gracefully.
60 requests per minute per key
Once your account is funded, the standard limit is 60 requests per minute per API key. This allows product teams to generate audio for assistive devices at a steady pace without immediate throttling.
10 requests per minute initially
New accounts start with a lower limit of 10 requests per minute. This allows you to test the integration without immediate strain on the system. After you add funds via USDT (TRC-20), your limit increases to the standard 60 requests per minute.
Real-time speed on long texts
Typical speed is roughly real time on long texts, generating about 25 characters of text per second of audio. Short phrases may take a few seconds to process. For texts exceeding 4,096 characters, split the input and join the resulting audio files.
Error Handling
When synthesis fails, the charge is returned in full. Failed requests are free, so you do not lose credits or character count for errors. This protects your budget during integration testing or network hiccups.
Failed requests are free
If the API cannot generate audio, you are not charged. The character count remains unchanged, and any prepaid amount is refunded immediately.
This applies to all failures, ensuring you only pay for successful audio generation.
- No deduction of characters on failure
- Full refund of charged amount
Standard HTTP error codes
The API returns standard HTTP status codes to indicate the result. Use 4xx codes for client errors like invalid keys or bad parameters, and 5xx for server issues.
Check the response body for specific error messages to resolve integration issues quickly.
- 401 for invalid API key
- 429 for rate limit exceeded
- 500 for server errors
Check status for synthesis issues
Monitor the response status to confirm if audio was generated successfully. For assistive devices, handle errors gracefully to avoid silent failures.
If a request fails, retry with exponential backoff or notify the user that audio is unavailable.
- Verify status code before processing audio
- Handle retries for transient errors

Code Examples
Start testing immediately with a simple curl command or integrate directly using the official OpenAI SDK. Both methods require only a minor configuration change to point to our API base URL.
Ready curl command for testing
Use this command to verify your key and generate speech instantly. Replace YOUR_API_KEY with your secret token and adjust the voice or model as needed.
OpenAI SDK configuration example
Leverage the existing OpenAI client libraries by updating the base URL. This allows you to use familiar methods like Audio.speech.create() without learning a new SDK structure.
Base URL adjustment for integration
Set your API client's base URL to https://api.playtk333.com/v1. This single change directs all requests to our TTS endpoint while maintaining standard request shapes.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Our assistive devices empower elderly users to navigate daily life with renewed confidence and independence through intuitive design.
23 languages
Any voice speaks any of these languages.
