Frequently asked questions
How do I get my API key?
Sign up on /get-api-key/ using Google or email. Your key appears immediately. It starts with 'tts-'. One active key per account. Issue a new key anytime to invalidate the old one instantly. Use it in the Authorization header as a Bearer token for all requests.
Can I use the OpenAI Python SDK with this API?
Yes. The endpoint matches the OpenAI speech shape. Change the base_url to https://api.getking11.com/v1 and use your site key. This allows direct compatibility with official OpenAI SDKs without writing custom client code, simplifying integration for your e-commerce text-to-speech workflows.
What is the maximum text length per request?
The input field accepts up to 4096 characters per request. For longer product descriptions, split your text into chunks of this size. Generate audio for each chunk separately, then join the resulting audio files to create the final output for your customers.
Does the API support streaming audio?
No. The API returns the complete audio file in the response body; there is no WebSocket or real-time partial audio streaming. You receive the full file once synthesis finishes. For long texts, generation typically takes roughly real-time duration (about 25 characters per second), though requests are processed sequentially and may queue under load.
How do I handle rate limit errors?
The limit is 60 requests per minute per key (10 before first top-up). If you exceed this, requests may queue or fail. Implement exponential backoff in your code. Monitor your usage dashboard to adjust batch sizes and avoid hitting the per-minute ceiling.
What happens if I exceed the character limit?
Requests exceeding 4096 characters will fail. To process longer texts, split your input into chunks of up to 4096 characters and join the resulting audio files. Failed requests are free, so you will not be charged for the oversized attempt.
Is the API key reusable across environments?
Yes. Use the same API key in development, staging, and production environments. One key is issued per account. You can revoke it and generate a new one at any time, which immediately invalidates the previous key across all environments.
Can I specify the language explicitly?
Yes. Pass the optional 'language' field with an ISO code. If omitted, the system guesses the language, defaulting to English for Latin script. Explicitly setting the language ensures correct pronunciation for non-English texts, such as German or Japanese, in your store.
First request
curl
curl https://api.getking11.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.getking11.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.getking11.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
Authentication for the LexiRead TTS API relies on a single active API key per account, issued via the /get-api-key/ dashboard. You must include this key in every request header to authorize usage and track character consumption against your balance.
API Key Format
Your API key is a string starting with tts- followed by alphanumeric characters. This key is unique to your account on getking11.com and serves as the sole identifier for billing and rate limit tracking.
Bearer Token Usage
Include the key in the HTTP header as Authorization: Bearer tts-.... Ensure no extra spaces or quotes surround the token value. The API rejects requests missing this header or containing invalid formats.
Key Rotation
You can issue a new key from the dashboard at any time. The previous key stops working immediately upon generation of the new one, ensuring no overlap or duplicate billing. Use this feature if your key is exposed or if you rotate credentials between development and production environments.
Endpoints Overview
The API exposes three core endpoints for generation and discovery. The speech endpoint handles audio synthesis, while the voices and models endpoints provide metadata about available options and engine versions. These endpoints use standard HTTP methods and JSON payloads, ensuring compatibility with existing developer workflows.
Speech Generation Endpoint
Send a POST request to /v1/audio/speech with your text, voice selection, and format preference. The API returns the synthesized audio file directly in the response body. This is the primary endpoint for converting product descriptions or notifications into audio.
Voices List Endpoint
Query GET /v1/voices to retrieve the list of available voice presets, supported languages, and output formats. This endpoint helps you verify which voices are active and which languages they support without guessing.
- Returns available voice presets
- Lists supported languages
- Shows available audio formats
Models Endpoint
Access GET /v1/models to view the current engine versions, such as tts-1 and tts-1-hd. Both models use the same engine and pricing, so you can choose based on your quality needs without worrying about tiered pricing structures.

Request Parameters
The API accepts standard request parameters to define audio output. You must specify a model identifier, select one of eight voice presets, and choose a response format. These fields control the synthesis engine and delivery method for your generated audio files.
Model Selection
Set model to either "tts-1" or "tts-1-hd". Both identifiers invoke the same underlying engine with identical pricing and characteristics. The choice is purely for identifier consistency with OpenAI-style clients, as there is no fidelity or speed difference between them.
Voice Presets
Select one of eight voices from aria, nova, luna, sage, felix, milo, atlas, or orion. Each voice is based on licensed LibriTTS-R recordings from LibriVox. These presets cover various tones suitable for product descriptions, from warm narration to energetic promotions.
- Female: aria, nova, luna, sage
- Male: felix, milo, atlas, orion
Response Formats
Specify response_format as mp3, wav, opus, flac, or pcm. MP3 is optimal for web delivery due to smaller file sizes, while WAV or FLAC may be preferred for high-quality archival storage. The API returns the audio file directly in the response body.
Rate Limits
Rate limits are enforced per API key to ensure stable processing. The system processes requests sequentially, meaning that under heavy load, a request may wait in a queue before synthesis begins.
Understanding these limits helps you plan integration workflows, especially when generating product descriptions in bulk or handling real-time notifications for your online store.
Standard Limit
After your first top-up, each API key allows 60 requests per minute. This limit applies uniformly regardless of whether you are using trial credit or paid balance.
Pre-Top-up Limit
Before the first top-up is completed, the limit is reduced to 10 requests per minute. This lower threshold remains in effect until a successful USDT (TRC-20) transfer is processed and credited to your account.
Concurrency Rules
Requests are processed one after another. If you send multiple requests rapidly, they will queue up. Because synthesis takes time (roughly real-time speed for long texts), high concurrency may increase total processing time. Plan your batch jobs accordingly to avoid unnecessary waiting periods.
Error Handling
Handling API failures is straightforward because the system guarantees that if synthesis fails, the charge is returned in full. This ensures you never pay for audio that was not delivered, simplifying cost tracking for high-volume e-commerce workflows.
Common Error Codes
The API returns standard HTTP status codes to indicate success or failure. A 200 OK response confirms the audio file is included in the response body. Non-200 codes, such as 400 or 429, indicate issues with the request or rate limits.
- 200: Request succeeded, audio file attached.
- 400: Invalid parameters, e.g., text too long or unknown voice.
- 401: Unauthorized, check your Bearer token.
- 429: Rate limit exceeded (60 req/min or 10 req/min pre-top-up).
Retry Logic
For transient network errors (5xx), implement exponential backoff. Retry up to 3 times before failing. For rate limits (429), wait the duration specified in the Retry-After header. Since requests are processed sequentially under load, brief pauses between retries prevent queue congestion.
- Retry on 5xx errors with exponential backoff.
- Respect Retry-After headers on 429 responses.
- Limit retries to 3 attempts to avoid excessive load.
Failed Request Refunds
If a request fails to produce audio, the full character count is refunded. The refund is processed automatically, so your balance reflects the credit immediately. No manual claim is needed. This applies to both failed synthesis and network errors that prevent audio delivery.
- Full refund for failed synthesis.
- Automatic credit restoration to account balance.
- No manual claim required.

OpenAI SDK Integration
Developers can integrate LexiRead’s TTS API using official OpenAI SDKs by simply overriding the base URL. This allows seamless adoption of familiar libraries without rewriting HTTP clients, accelerating integration for product description generation and automated notifications.
Base URL Configuration
Point your SDK’s base_url to https://api.getking11.com/v1. The API follows the same request shape as the OpenAI Speech endpoint, ensuring that standard SDK methods work without modification to the core logic.
SDK Compatibility
The official OpenAI SDKs are fully compatible. You only need to adjust the base URL and inject your tts-... bearer token. No custom client logic is required, as the endpoint structure mirrors the standard OpenAI speech API.
Code Snippets
Send a POST request with your model, input text, and voice selection. The response returns the complete audio file in the body, not a stream. This ensures reliable delivery for static assets like product descriptions, though it may introduce slight latency for real-time notifications compared to streaming APIs.
Text Splitting Strategy
The API enforces a strict 4096-character limit per request. To process longer texts, split your input into sequential chunks and send them as individual POST requests. Each chunk must be treated as an independent synthesis job, as the service does not maintain state between calls.
The client is responsible for retrieving each resulting audio file and concatenating them in order. Because the API returns discrete files, the playback continuity depends entirely on your implementation of the joining logic. Gaps or minor discontinuities may occur at the boundaries depending on how your code handles the transition between files.
4096 Character Limit
Send text in blocks of up to 4096 characters. Spaces and punctuation count toward this limit. If your product description or notification exceeds this, divide it into sequential parts before sending them to the endpoint.
Each part generates a separate audio file. Ensure your application tracks the order of requests to maintain the correct sequence of the final audio output.
- Limit: 4096 characters per request
Audio Joining Logic
The API returns distinct audio files for each chunk. Your code must download these files and join them sequentially. The service does not guarantee seamless playback at the join points; any gaps or discontinuities depend on how your client handles the concatenation.
Use a reliable audio library to merge the files. Ensure the output format matches the requested response_format to avoid compatibility issues during the merge process.
Performance Impact
Processing time scales with the number of chunks. Short phrases take a few seconds, while longer texts generate at roughly real-time speed (25 characters per second). Requests are processed sequentially, so under load, a request may wait in a queue.
Splitting large texts into multiple requests increases latency due to separate HTTP calls and joining overhead. Plan your splitting strategy to balance between request count and integration complexity.
- Speed: ~25 chars/sec for long texts

Language Detection
Language handling determines how the API interprets your input text. The system guesses the language from the script and defaults to English for Latin text. This ensures correct pronunciation without requiring explicit configuration for every request.
Auto-Detection Rules
The API infers the language based on the character set in your input. If you provide non-Latin characters, the system identifies the script to determine the appropriate language model.
- Latin script defaults to English if no language is specified.
- Non-Latin scripts trigger automatic language identification based on the character set.
Manual Language Code
For non-English Latin-script text, you must explicitly set the language field. This prevents the API from defaulting to English and ensures accurate pronunciation for languages like French, German, or Spanish.
- Use the
languageparameter with a valid ISO code. - Applies to all 23 supported languages regardless of script.
Default Language Fallback
If the language field is omitted, the system assumes English for Latin text. For other scripts, it guesses the language based on the characters. Unsupported languages will not be processed correctly.
- English is the default for unspecified Latin text.
- Non-Latin text relies on script-based guessing.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Build your online stores to scale quickly and attract new customers.
23 languages
Any voice speaks any of these languages.
