Frequently asked questions
How do I get an API key?
Visit /get-api-key/ to sign up with Google or an email and password. Upon authentication, the dashboard displays your unique API key, current balance, and usage statistics. You can issue a new key at any time, which immediately invalidates the previous one.
Can I use the OpenAI SDK?
Yes. The endpoint shape matches the OpenAI speech endpoint. Update your SDK's base_url to https://api.playbe99.com/v1 and use your site-specific key. This allows you to use official OpenAI SDKs without custom client implementations for standard text-to-speech requests.
What is the base URL?
The API base URL is https://api.playbe99.com/v1. All requests, including speech synthesis and voice listing, are directed to this domain. Ensure your client configuration points to this address to avoid routing errors when sending text for audio generation.
Which audio formats are supported?
Supported formats are mp3, wav, opus, flac, and pcm. Specify the desired format in the response_format field. These options cover standard playback needs for iOS and Android applications, ensuring compatibility with most mobile audio players and processing pipelines.
What are the rate limits?
The limit is 60 requests per minute per key. New accounts without top-ups are limited to 10 requests per minute. Exceeding these limits will return errors.
How do I handle errors?
Failed requests are free, and the charge is returned in full. Monitor your response status codes to detect synthesis failures. Since requests are processed sequentially, high load may cause delays. Check your dashboard for usage details and balance status after errors.
Can I change my API key?
Yes, you can issue a new key at any time via the dashboard. When a new key is generated, the old one stops working immediately. This allows you to rotate credentials securely without downtime, provided your client updates to the new key.
Does the API support streaming?
No streaming or WebSocket support is available. You must send a POST request and receive the complete audio file in the response. This synchronous approach simplifies integration for mobile apps that do not require real-time partial audio generation.
Is SSML supported?
SSML is not supported. You cannot use phoneme dictionaries or custom pronunciation lexicons. Input text is processed directly, so ensure your strings are correctly formatted before sending them to the endpoint to avoid unexpected pronunciation issues in the generated audio.
How do I split long texts?
Split input text into parts of up to 4096 characters. Generate audio for each part separately, then join the resulting audio files in your application. This ensures compliance with the character limit per request while allowing for longer content generation.
What is the character limit?
Each request supports up to 4096 characters. Spaces and punctuation count toward this limit. For texts exceeding this length, you must split the input manually or via your client logic and process each segment individually to ensure successful synthesis.
Can I adjust the speed?
Yes, the speed parameter accepts values between 0.25 and 4.0. Adjusting speed affects the audio playback rate. This feature allows you to optimize for faster processing or specific narrative pacing requirements within your mobile application context.
How do I select a voice?
Choose from eight presets: aria, nova, luna, sage, felix, milo, atlas, or orion. Each voice supports all 23 languages. Select the voice that best matches your app's tone, ensuring the language parameter is set correctly for non-English Latin-script text.
What happens if I exceed limits?
Requests are processed sequentially and may wait in a queue under load. If you exceed the rate limit, requests may fail or delay. Failed requests incur no charge. Monitor your dashboard to track usage and adjust your client's request frequency accordingly.
First request
curl
curl https://api.playbe99.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.playbe99.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.playbe99.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 and API Keys
Access the /get-api-key/ endpoint to create an account via Google or email credentials. This step initializes your dashboard, displaying your unique API key, current balance, and usage metrics specific to playbe99.com.
Signing Up for an Account
Register using Google for immediate access or email/password for standard accounts. Google sign-ins receive a $0.50 trial credit valid for seven days, providing approximately 50,000 characters for testing. Email-based accounts start with zero balance and require manual top-up.
Generating Your Bearer Token
Upon account creation, a single active API key is generated. Use this key in the Authorization header as Bearer tts-... for all requests. The key is visible immediately on your dashboard alongside your character count and current balance.
Rotating Keys Instantly
Generate a new key at any time from the dashboard. The previous key becomes inactive immediately upon issuance of the new one, ensuring zero downtime if you switch keys programmatically. Only one key is active per account at any given moment.
First Request with cURL
To integrate Voxera’s text-to-speech engine, send a POST request to https://api.playbe99.com/v1/audio/speech. The endpoint accepts JSON payloads containing the input text and voice configuration. This structure mirrors standard TTS API conventions, ensuring straightforward implementation in iOS and Android environments.
The request requires a valid bearer token in the header and a body defining the model, input text, voice preset, and output format. Responses return the generated audio file directly, allowing immediate playback or storage within your mobile application’s asset directory.
POST Request Structure
Direct your HTTP client to the speech endpoint using the POST method. Include your API key in the Authorization header as Bearer <your-key>. The request body must be formatted as JSON, containing only the required parameters for synthesis.
JSON Body Parameters
The JSON body defines the synthesis parameters. Key fields include model (selecting between tts-1 or tts-1-hd), input (the text string up to 4096 characters), voice (one of the eight available presets), and response_format (mp3, wav, opus, flac, or pcm). Optional fields allow speed adjustment and language specification.
Ensure the input text does not exceed 4096 characters. If your content is longer, split it into chunks and concatenate the resulting audio files. The language field is optional but recommended for non-English Latin-script texts to prevent incorrect language guessing.
- model: "tts-1" or "tts-1-hd"
- input: string (max 4096 chars)
- voice: e.g., "aria", "nova"
- response_format: "mp3", "wav", "opus", "flac", or "pcm"
- speed: float (0.25–4.0)
Response Handling
A successful response returns the audio file directly in the response body with the appropriate Content-Type header (e.g., audio/mpeg for MP3). Save this binary data to a local file or stream it directly to your audio player.
If synthesis fails, the response includes an error object. Failed requests incur no charge, and the billing adjustment is automatic. Check the HTTP status code to determine if retry logic is needed, keeping in mind the rate limit of 60 requests per minute.
OpenAI SDK Integration
Voxera’s API mirrors the OpenAI speech endpoint structure, allowing mobile developers to reuse existing client libraries with minimal configuration. By simply updating the base URL to https://api.playbe99.com/v1, you can leverage official OpenAI SDKs for seamless integration into iOS and Android applications without learning a new protocol.
Changing the Base URL
Point your SDK client to https://api.playbe99.com/v1. This single change redirects all requests to our infrastructure while maintaining standard request shapes. No custom parsers or middleware are required.
Using Official OpenAI SDKs
Use the standard OpenAI SDK for your preferred language. Pass your Bearer token via the Authorization header. The SDK handles JSON serialization and response parsing exactly as it would for OpenAI endpoints, ensuring consistent behavior across platforms.
- Supports Python, Node.js, Go, and Java clients.
Voice and Model Selection
Select from eight voice presets and two models (tts-1 or tts-1-hd). The model parameter does not affect price or engine quality, and both models use the same engine. Specify language codes explicitly for non-English Latin text to ensure accurate synthesis.
- tts-1 and tts-1-hd use the same engine.
Available Audio Formats
Selecting the correct audio format is critical for iOS and Android app performance, balancing file size against decoding overhead. The API supports five distinct response formats, each optimized for different deployment scenarios. Developers must explicitly declare their preferred format in the request body to ensure consistent delivery.
MP3 for Compact Storage
MP3 remains the standard for mobile distribution due to its universal compatibility and small footprint. It is ideal for caching audio assets on devices with limited storage. This format reduces bandwidth costs while maintaining acceptable fidelity for most speech content.
- Universal support across iOS and Android versions.
- Smallest file size among lossy formats.
WAV for High Quality
WAV provides uncompressed, raw audio data, ensuring zero quality loss during transmission. Use this format when precise audio analysis or high-fidelity playback is required. It is suitable for short phrases or notifications where file size is less critical than immediate, pristine playback.
- Uncompressed PCM audio data.
- Zero quality loss for critical audio segments.
- Simpler decoding for immediate playback.
Opus, FLAC, and PCM Options
Opus offers high compression efficiency. FLAC provides lossless compression, reducing storage needs without sacrificing quality. PCM delivers raw linear audio, useful for direct DSP processing. Choose based on your app’s specific audio pipeline requirements.
- Opus: High compression efficiency.
- FLAC: Lossless compression for archival.
- PCM: Raw data for direct processing.
Voice Presets and Models
Voxera provides eight distinct voice presets optimized for mobile app contexts, ranging from warm narrators to energetic promotional voices. These presets are based on licensed recordings from LibriTTS-R and LibriVox public-domain audiobooks, ensuring clear licensing clarity for iOS and Android deployments. Each voice supports all 23 available languages without requiring specific model selection.
Selecting from 8 Voice Presets
The available voices include female options like Aria (warm, clear narrator) and Nova (bright, upbeat), as well as Luna and Sage for storytelling and instructional content. Male presets include Felix for energetic promos, Milo for conversational tones, Atlas for neutral news delivery, and Orion for deep documentary styles. These presets are uniform across all models.
- Aria, Nova, Luna, Sage (Female)
- Felix, Milo, Atlas, Orion (Male)
Using tts-1 and tts-1-hd Models
Both "tts-1" and "tts-1-hd" utilize the same underlying engine and pricing structure. The choice between them does not affect the voice preset or language capabilities, as the audio generation remains identical. Developers can select either model string in the request body without altering the output quality or character cost.
Language Parameter Specification
While any voice can speak any of the 23 supported languages, you should explicitly set the language parameter for non-English Latin-script text. If omitted, the system guesses the language, defaulting to English for Latin scripts. This ensures accurate pronunciation for languages like Dutch, German, or Swedish, avoiding misinterpretation by the auto-detection logic.

Rate Limits and Throttling
Rate limits protect API stability during peak usage. Each API key allows a maximum of 60 requests per minute. This limit applies uniformly across all endpoints and voice models.
New accounts start with a lower threshold of 10 requests per minute. This baseline increases automatically after the first top-up, ensuring smooth integration during early development phases.
60 Requests Per Minute Limit
The standard rate limit is 60 requests per minute per key. This capacity supports typical mobile app workflows. Exceeding this limit triggers an HTTP 429 error.
- Limit applies per API key, not per user session.
- Limits reset on a sliding window basis.
- Exceeding limits returns a 429 status code.
10 Requests Per Minute Initially
New keys begin with a 10 requests per minute limit. This constraint remains active until you perform your first top-up. Once a minimum of $5 in USDT (TRC-20) is transferred, the limit automatically increases to the standard 60 requests per minute.
- Initial limit: 10 requests/minute.
- Increases after first successful top-up.
- No manual activation required.
Handling Queue Waits
Requests are processed sequentially during high load. If the queue fills, your request may experience delays. Typical generation speed is roughly one second of audio per 25 characters of text. Plan your retry logic to handle transient delays gracefully without duplicating charges.
- Requests process in order of arrival.
- Delays are normal under heavy load.
- Failed requests do not incur charges.
Error Handling and Status Codes
The API returns standard HTTP status codes to indicate request outcomes. A 200 OK confirms successful audio synthesis, while 400 Bad Request indicates invalid parameters such as an unsupported voice or model. Use 429 Too Many Requests when you exceed the 60 requests per minute limit or the initial 10 requests per minute threshold for new accounts.
Common HTTP Error Codes
Identify issues by inspecting the response body for specific error messages. Common codes include 401 Unauthorized for invalid or expired keys, 422 Unprocessable Entity for validation errors, and 500 Internal Server Error for temporary service disruptions during high load.
- 400 Bad Request: Invalid input or format.
- 401 Unauthorized: Invalid API key.
- 429 Too Many Requests: Rate limit exceeded.
- 500 Internal Server Error: Service temporary failure.
Retrying Failed Requests
Implement exponential backoff for 429 and 500 errors to manage queue waits during peak load. Since requests are processed sequentially, retries should account for potential queuing delays. Retry logic should be implemented in your mobile app’s network layer to ensure reliability without overwhelming the endpoint.
Billing Adjustments on Failure
If synthesis fails entirely, the charge is returned in full. Failed requests do not incur costs, ensuring you only pay for successfully generated audio files. Monitor your balance and usage logs via the dashboard to verify credits for any disputed failures.
Advanced Usage Patterns
Mobile applications often handle texts exceeding the 4,096-character limit. You must split input text into smaller chunks, send requests, and concatenate the resulting audio files.
Splitting Long Texts
Divide your input string at natural boundaries, such as sentence ends, ensuring no single request exceeds 4,096 characters. This prevents truncation and maintains voice consistency across segments.
- Limit input to 4,096 characters per request.
- Split at logical sentence or phrase boundaries.
- Maintain consistent voice and speed settings across chunks.
Joining Audio Files
Receive audio segments in your chosen format (mp3, wav, opus, flac, or pcm). Concatenate these files sequentially on the client side or server side before playback.
- Support for mp3, wav, opus, flac, and pcm.
- Sequential concatenation preserves audio continuity.
- No server-side merging required; handle locally.
Speed Adjustment Parameters
Adjust playback speed between 0.25x and 4.0x using the speed parameter. This allows optimization for character limits or tailored listening experiences without re-generating audio, provided the voice engine supports the specific speed setting.
- Range: 0.25 to 4.0.
- Optimize for data or listening preference.
- Applies to tts-1 and tts-1-hd models.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Create intuitive mobile applications that empower freelancers to work efficiently from anywhere in the world.
23 languages
Any voice speaks any of these languages.
