Frequently asked questions
Do I need the OpenAI SDK?
No, you can use any HTTP client. The API accepts standard JSON requests, so plain HTTP works perfectly. However, the official OpenAI SDKs work seamlessly by simply changing the base_url to https://api.start11.info/v1 and using your ToonVoce key. This allows you to leverage existing code structures without needing a proprietary SDK.
Can I use this with Python?
Yes, Python is fully supported. You can use the official OpenAI Python library by configuring the base_url to https://api.start11.info/v1. Alternatively, standard libraries like requests work directly with the JSON endpoint. Since the request shape matches the OpenAI speech endpoint, your existing Python scripts for audio generation will integrate easily.
How do I set the base URL?
Set your API base URL to https://api.start11.info/v1. This is required for the official OpenAI SDKs to route requests correctly. For direct HTTP calls, use the full endpoint https://api.start11.info/v1/audio/speech. Ensure your authentication header points to this domain to avoid routing errors and ensure your requests reach the correct service.
What happens if I exceed the rate limit?
The limit is 60 requests per minute per key (10 per minute before the first top-up). If you exceed this, new requests will queue and process sequentially. Requests are handled one after another, so you may experience delays during high load. Failed requests remain free, but successful ones consume your character balance according to the standard pricing model.
Can I stream audio in real-time?
No, streaming is not supported. The API returns the complete audio file in the response body after synthesis finishes. You must wait for the full generation to complete before receiving the data. This means you cannot receive partial audio chunks in real-time; you get the entire file once processing is done.
Is SSML supported for emphasis?
No, SSML is not supported. You cannot use SSML tags for emphasis, pauses, or pronunciation adjustments. The API relies on the text input and voice presets to determine delivery. If you need specific pacing or emphasis, you must manage that logic in your application or choose a voice preset that matches the desired tone.
How do I specify a language?
Include the language field in your JSON body with an ISO code, such as en, de, or ja. If omitted, the system guesses the language from the script, defaulting to English for Latin text. For non-English Latin-script text, explicitly setting the language field ensures accurate pronunciation and avoids misinterpretation by the engine.
What is the max character limit?
Each request supports up to 4096 characters in the input field. Spaces and punctuation count toward this limit. If your text exceeds this length, you must split it into smaller chunks. Each chunk generates a separate audio file, which you can then join together in your application to create the final continuous audio track.
Do I need to split long texts manually?
Yes, you must split texts longer than 4096 characters yourself. The API does not handle automatic splitting. You should divide your text into parts, send each part in a separate request, and then join the resulting audio files. This ensures each request stays within the character limit and generates valid audio for each segment.
Are requests processed in parallel?
No, requests are processed sequentially. Under load, a request may wait in a queue before synthesis begins. This means processing time depends on the queue length. For long texts, generation takes roughly real-time speed, but short phrases may take a few seconds. Parallel processing is not available, so plan your workflow accordingly.
Can I change the voice mid-sentence?
No, you cannot change the voice mid-sentence within a single request. Each request generates audio for one specific voice. To switch voices, you must split the text at the sentence boundary, send separate requests for each voice, and then join the audio files. This allows for character changes between distinct audio segments.
Is the API RESTful?
Yes, the API is RESTful. It uses standard HTTP methods, specifically POST for speech generation and GET for retrieving voices and models. Requests are sent as JSON, and responses return audio files or metadata. This standard structure allows easy integration with any platform that supports RESTful HTTP clients and JSON data exchange.
How do I handle audio format conversion?
Specify the desired format using the response_format field: mp3, wav, opus, flac, or pcm. The API returns the audio file in that exact format. No additional conversion is needed on your end. If you require a different format, you can convert the received file locally using standard audio tools after downloading the response.
What is the authorization header format?
Use the header Authorization: Bearer tts-... with your unique API key. Each account has one active key at a time. If you issue a new key, the old one stops immediately. Ensure you include this header in every request to authenticate your account and track usage against your balance correctly for billing purposes.
First request
curl
curl https://api.start11.info/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.start11.info/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.start11.info/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
ToonVoce uses a single active API key per account for all requests. You can generate or revoke keys directly from the /get-api-key/ dashboard. Each key is tied to your specific site account and tracks usage independently.
API Key Management
Keys are issued when you sign up via Google or email. The dashboard displays your current key, balance, and character usage. You can create a new key at any time to replace the old one.
- One active key per account
- Instant invalidation of old key upon new issuance
- View key and balance on /get-api-key/
Key Rotation Process
Rotate keys by generating a new one in the dashboard. The old key stops working immediately. Update your scripts or SDK configuration to use the new key before the old one expires or is revoked.
- Generate new key in dashboard
- Old key invalidates instantly
- Update base URL config if needed
Authorization Header Format
Include your key in the HTTP header for every request. Use the format: Authorization: Bearer tts-<your-key>. Ensure no extra spaces or characters are included in the token string.
- Header: Authorization: Bearer tts-<key>
- Case-sensitive bearer token
- Required for all API endpoints
Base URL Configuration
The API base URL is https://api.start11.info/v1. All requests use POST to /audio/speech, mirroring the standard OpenAI speech endpoint shape.
Endpoint Structure
Send your text input as JSON to this endpoint. The request body requires the model, input, voice, and response_format fields.
Versioning Details
The /v1 prefix ensures consistent behavior. We do not version the base URL dynamically; the endpoint structure remains stable.
SDK Compatibility
Since the request shape matches OpenAI’s specification, you can use the official OpenAI SDKs by setting the base_url to our API address.
- No custom SDK required
- Works with standard HTTP clients
First Request with cURL
Get your first audio file in seconds. Send a POST request to the speech endpoint with your API key and a text payload under 4,096 characters. The API returns the audio file directly in the response body, ready to play or save.
Use the official OpenAI SDKs by pointing them to our base URL, or send a raw HTTP request. This section shows the simplest cURL command to generate speech from a script draft.
Sending Text Input
Construct a JSON body containing the model (e.g., tts-1), voice (e.g., nova), and input text. Include response_format (mp3, wav, etc.) and optional speed or language. Ensure the text does not exceed 4,096 characters per request.
For longer scripts, split the text into chunks and join the resulting audio files. This ensures consistent quality and avoids truncation.
- Use
POST https://api.start11.info/v1/audio/speech - Set
Authorization: Bearer tts-...header - Include
model,voice,input, andresponse_format
Response Format
On success, the API returns the audio file directly in the response body with the appropriate content type (e.g., audio/mpeg). The file is ready for immediate playback or storage in your animation pipeline.
No additional parsing is needed for the audio data; it is returned as a binary stream or base64-encoded data depending on the client library used.
Error Handling Basics
If synthesis fails, the request is free, and you receive a JSON error object with a descriptive message. Check the status code and message field to identify issues like invalid voice names or character limits.
Handle HTTP 400 errors for invalid input and 401 errors for authentication problems. Retrying failed requests is safe since they do not consume credits.
Supported Voices
ToonVoce provides eight preset voices trained on licensed recordings, ensuring consistent character performance across 23 languages. Each voice is derived from LibriTTS-R or LibriVox public-domain sources, allowing agencies to use them in commercial animation drafts without additional licensing hurdles.
Voice Descriptions
Our library offers distinct vocal personalities optimized for animation workflows. Female presets include Aria (warm narrator), Nova (bright and upbeat), Luna (soft storyteller), and Sage (calm instructor). Male options feature Felix (energetic promo), Milo (friendly conversational), Atlas (neutral news), and Orion (deep documentary).
- Aria: Warm, clear narrator
- Nova: Bright, upbeat
- Luna: Soft, storytelling
- Sage: Calm, instructional
- Felix: Energetic, promo
- Milo: Friendly, conversational
- Atlas: Neutral, news
- Orion: Deep, documentary
Gender and Tone Options
Select voices based on tonal requirements rather than strict gender roles, as any voice can speak any supported language. Use Felix for high-energy trailers or Orion for serious documentary segments. Mix voices within a project to distinguish characters without hiring additional talent.
Voice Selection Guide
Match the voice to the script’s emotional core. For children’s content, Luna or Milo provide approachable tones. Use Atlas for factual exposition or Aria for narrative-driven scenes. All voices support the full range of 23 languages, so selection depends on tone, not linguistic constraints.
Audio Formats
ToonVoce supports five standard audio formats to suit different animation pipelines. You can choose between lossy and lossless options depending on your quality requirements.
PCM Format Details
PCM provides uncompressed, raw audio data. It is ideal for direct processing in professional editing software without additional decoding steps.
MP3 and WAV Options
MP3 offers efficient compression for web delivery, while WAV provides high-quality uncompressed audio suitable for broadcast standards.
Opus and FLAC Support
Opus delivers excellent quality at low bitrates. FLAC offers lossless compression, preserving all audio details without file size bloat.
Speed and Language
Controlling pacing and ensuring correct language detection are key to consistent character voices in animation. Our API handles language guessing automatically but allows precise overrides when needed.
Speed adjustments let you match the energy of a scene, while language codes ensure accurate pronunciation across the 23 supported languages.
Latin Script Defaults
If the language field is omitted, the system guesses the language from the text script. Latin-script text defaults to English. For non-English Latin-script dialogue (e.g., Spanish or German), always set the language field to avoid incorrect pronunciation.
Speed Adjustment Range
Adjust speech tempo with the speed parameter, accepting values from 0.25 to 4.0. A speed of 1.0 is normal. Use faster speeds for energetic promos or slower speeds for storytelling. Note that generation time scales with speed; faster requests may process quicker but still follow queue order.
Language Code Specification
Specify the ISO language code (e.g., en, de, ja) for precise control. Any voice can speak any of the 23 supported languages. Unsupported languages like Vietnamese or Thai will not work correctly even if passed.
- Arabic (ar), Chinese (zh), Danish (da), Dutch (nl), English (en), Finnish (fi), French (fr), German (de), Greek (el), Hebrew (he), Hindi (hi), Italian (it), Japanese (ja), Korean (ko), Malay (ms), Norwegian (no), Polish (pl), Portuguese (pt), Russian (ru), Spanish (es), Swahili (sw), Swedish (sv), Turkish (tr)

Rate Limits and Errors
Rate limits protect the service during peak animation production hours. Each API key allows 60 requests per minute once your account is funded, with a lower 10 requests per minute limit before your first top-up.
Exceeding these limits triggers a 429 error. The system processes requests sequentially, so under heavy load, synthesis may queue briefly. Failed requests incur no charge, ensuring you only pay for successfully generated audio.
Handling 429 Errors
When you send too many requests, the API returns a 429 status code. Implement exponential backoff in your animation pipeline to pause briefly before retrying. This prevents overwhelming the server during large-scale voiceover drafts.
Request Queue Behavior
Requests are processed one after another. If multiple scripts are sent simultaneously, they enter a queue. Generation speed is roughly real-time, but short phrases may take a few seconds. Patience ensures smooth integration with your studio’s workflow.
Character Limit Rules
Each request accepts up to 4096 characters, including spaces and punctuation. For longer scripts, split the text into chunks, generate each part, and join the audio files. This ensures consistent quality and avoids exceeding the single-request limit.
Voices and Models Endpoint
The API provides programmatic access to voice configurations and available models, allowing developers to dynamically select the best voice for their animation project. This endpoint returns a comprehensive list of supported voices, their associated metadata, and the current model status.
Response Structure
The GET /v1/voices endpoint returns a JSON object containing an array of voice objects. Each object includes the voice ID, gender, description, and supported languages.
- Voice ID (e.g., aria, nova)
- Gender (female/male)
- Description (e.g., warm, clear narrator)
- Supported languages list
- Available response formats
Model Selection
The GET /v1/models endpoint returns the current model identifiers, specifically "tts-1" and "tts-1-hd". Both models use the same underlying engine and pricing. The response confirms the model status and ensures compatibility with your SDK configuration.
Available Languages List
The API supports 23 languages, including Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish. Any voice can speak any of these languages, providing consistent quality across your animation projects regardless of the target locale.
Pricing and Payment
ToonVoce charges $10 per million characters, which equals $0.01 per 1,000 characters. Spaces and punctuation count toward this limit. Failed synthesis requests are free, with charges returned in full if generation fails.
Cost Per Character
Usage is billed strictly by character count, including spaces and punctuation. The rate is $10 per 1,000,000 characters ($0.01 per 1,000 characters). There are no monthly plans, seats, or minimum spend requirements.
Free Trial Details
New accounts created via Google sign-in receive a $0.50 free trial credit, valid for 7 days (approx. 50,000 characters). Accounts created with email and password start without trial credit.
Top-Up Methods
Balance is topped up via USDT on the TRON network (TRC-20) only. The minimum top-up is $5. Credits are credited automatically within minutes of transfer. There are no credit cards, PayPal, subscriptions, or invoices with VAT.
- USDT (TRC-20) on TRON network only
- Minimum top-up: $5
- Automatic credit within minutes
- No cards, PayPal, or VAT invoices
Account Management
Create your ToonVoce account on the /get-api-key/ page using either Google sign-in or a standard email and password. Each account is specific to this site, providing you with immediate access to the API dashboard and your unique authentication credentials.
Once registered, you can manage your credentials securely. The platform supports a simple key rotation process where issuing a new key immediately invalidates the old one, ensuring your production pipelines remain secure without complex reconfiguration.
Sign Up Process
Registration is straightforward: choose Google OAuth for quick access or create a traditional account with email and password. Note that accounts created via email do not automatically receive the $0.50 free trial credit, which is reserved for Google sign-ins. Your account is isolated to this storefront, ensuring dedicated management of your API key and usage.
Balance Tracking
Monitor your credit usage in real-time via the dashboard. The interface displays your current balance, remaining trial credit, and total characters processed. Since pricing is $0.01 per 1,000 characters, you can accurately predict costs for your animation drafts. Failed requests are free, so your balance only reflects successful synthesis.
- Real-time balance updates
- Clear character count tracking
- Instant refund for failed requests
Usage Statistics
View detailed logs of your API activity, including timestamps, character counts, and status codes. This data helps you optimize your workflow for voice-over drafts without needing external tools. Request metadata is retained for billing purposes but input text is not used for training.
Data Privacy
ToonVoce uses your input text exclusively to generate the requested audio file. We do not use your scripts, dialogue, or narration data to train our models or improve the voice engines, ensuring your creative content remains yours.
Input Text Usage
Your text is processed solely for synthesis. Once the audio is generated and delivered, the input text is not retained for training purposes.
Audio Storage Policy
Generated audio files are returned in the API response and are not stored on our servers. If you need to keep the audio, you must save it to your own infrastructure.
Billing Metadata
We retain request metadata, including timestamp, character count, and status, strictly for billing accuracy and usage tracking. This data helps us process your top-ups and monitor rate limits.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Each voice has a distinct tone suitable for standard adult roles in animation.
23 languages
Any voice speaks any of these languages.
Sources and further reading
- Wikipedia
- Tether (USDT)
- openai-node on GitHub
- openai-python on GitHub
- OpenAI API reference: create speech
- IETF RFC 7845: Ogg Opus
- Python documentation
- LibriTTS-R paper (arXiv 2305.18802)
- LibriTTS-R corpus (OpenSLR 141)
- curl documentation
- W3C WAI: making audio and video accessible
- LibriVox public-domain audiobooks
