Frequently asked questions
How do I get an API key?
Navigate to /get-api-key/ and sign in with Google or create an account using an email and password. The dashboard displays your active API key, current balance, and usage statistics. Google sign-ins receive a $0.50 trial credit valid for seven days, while email accounts start without credit.
Which SDKs are supported?
The API follows the same request shape as the OpenAI speech endpoint. You can use the official OpenAI SDKs by changing the base_url to https://api.368g.info/v1. Alternatively, you may use any standard HTTP client to send POST requests with the required JSON body and authentication headers.
What is the base URL?
The API base URL is https://api.368g.info/v1. You send POST requests to https://api.368g.info/v1/audio/speech for synthesis. Additional endpoints include GET /v1/voices for voice details and GET /v1/models for model information. Ensure your SDK configuration points to this domain for correct routing.
How do I set the language?
Include the optional 'language' field in your request body with a valid ISO code. If omitted, the system guesses the language from the script, defaulting to English for Latin text. Explicitly setting the language is recommended for non-English Latin-script text to ensure accurate synthesis.
What is the character limit?
Each request accepts up to 4096 characters of input text. For longer texts, split the content into parts of this size and join the resulting audio files. Spaces and punctuation count toward the character limit, which is also used for billing calculations at $0.01 per 1,000 characters.
Does it support streaming?
No, streaming is not supported. The API returns the complete audio file in the response after synthesis finishes. Requests are processed sequentially, so under load, a request may wait in a queue. You must wait for the full response before receiving the audio data.
Can I use SSML?
No, SSML is not supported. The API accepts plain text input only. It does not offer phoneme dictionaries, custom pronunciation lexicons, or emotion tags. You must manage pacing and emphasis by splitting text into separate requests or adjusting the overall speed parameter within the allowed range.
How do I handle rate limits?
The limit is 60 requests per minute per key, or 10 per minute before the first top-up. Requests are processed sequentially, so queueing may occur under load. Monitor your usage via the dashboard. If you exceed limits, wait for the window to reset or upgrade your balance.
What happens on error?
If synthesis fails, the charge is returned in full. Failed requests do not deduct from your balance. You can retry the request or adjust parameters like speed or voice. Check the response status to identify specific errors, such as invalid input or rate limit exceedance.
Is the response audio stored?
No, the generated audio is returned in the response and is not stored by the service. Input text is used only to produce the audio and is not used for training. Request metadata, including time, characters, and status, is retained solely for billing purposes.
How do I change voices?
Specify the 'voice' field in your request body with one of the eight available presets. Options include aria, nova, luna, sage, felix, milo, atlas, and orion. Each voice supports all 23 languages. You can switch voices dynamically between requests without any additional configuration or setup.
Can I adjust speed dynamically?
Yes, use the 'speed' parameter in your request body. The acceptable range is 0.25 to 4.0. This allows you to adjust playback speed per request. Typical speed is roughly real time on long texts (about 25 characters of text per second of generation).
What formats are available?
You can request audio in mp3, wav, opus, flac, or pcm formats. Specify the desired format using the 'response_format' field in your request. All voices support all these formats. The choice affects file size and compatibility but not the synthesis engine or pricing structure.
How do I rotate my key?
Issue a new API key from your account dashboard at any time. When a new key is generated, the previous key stops working immediately. This ensures secure access control. You do not need to wait for expiration. Update your applications to use the new key promptly.
First request
curl
curl https://api.368g.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.368g.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.368g.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 and Setup
Access the VoiceCompliance API by generating a unique key via the /get-api-key/ dashboard. Each account holds one active key; include it in the Authorization header as a Bearer token to authenticate requests.
API Key Generation
Keys are issued upon sign-in via Google or email. The dashboard displays the current key and usage statistics. A new key can be generated at any time, instantly invalidating the previous one.
- One active key per account
- Instant invalidation upon rotation
- Viewable in dashboard
Bearer Token Usage
Include the header Authorization: Bearer tts-<your-key> in every POST request to /v1/audio/speech. Ensure no extra whitespace surrounds the token value to prevent authentication failures.
Key Rotation Policy
Rotate keys immediately if compromised. The old key stops working at once upon issuance of the new one. Update your integration configuration before the old key expires from active use.
Endpoint Specifications
The VoiceCompliance API provides three primary endpoints for managing voice-over generation: speech synthesis, voice listing, and model information. The Speech Generation endpoint accepts text and configuration parameters to return audio files. The Voice Listing endpoint returns available presets, supported languages, and format options. The Model Information endpoint details the underlying engine specifications.
Speech Generation POST
Send text to POST https://api.368g.info/v1/audio/speech to receive an audio file. The request body uses the standard OpenAI speech endpoint structure. Input text is limited to 4096 characters per request. Failed requests incur no charge.
- Endpoint: POST /v1/audio/speech
- Body: JSON with model, input, voice, response_format, speed
Voice Listing GET
Retrieve available voices, languages, formats, and pricing via GET https://api.368g.info/v1/voices. This endpoint confirms the 23 supported languages and 8 voice presets. All voices can speak any supported language.
- Endpoint: GET /v1/voices
- Returns: voice_id, language, format, price
Model Information GET
Access model details at GET https://api.368g.info/v1/models. Both tts-1 and tts-1-hd use the same engine and price. This endpoint confirms the model capabilities without requiring authentication.
- Endpoint: GET /v1/models
- Models: tts-1, tts-1-hd
Request Parameters
The API accepts plain text input with a strict limit of 4096 characters per request. Spaces, punctuation, and special characters count toward this total. For longer content, split the text into multiple requests and join the resulting audio files sequentially to ensure consistent playback.
Input Text Limit
Each request must not exceed 4096 characters. If your script exceeds this length, divide it into chunks and concatenate the output files. This ensures reliable synthesis without truncation.
Voice Selection
Select from eight preset voices (aria, nova, luna, sage, felix, milo, atlas, orion). Each voice supports all 23 supported languages. Specify the voice name in the voice field. If the language parameter is omitted, the system guesses the language, defaulting to English for Latin scripts.
Set the language parameter explicitly for non-English Latin text to ensure accurate pronunciation. All voices are licensed from LibriTTS-R (CC BY 4.0) and LibriVox public-domain readings.
Speed Control Range
Adjust the playback speed using the speed parameter, which accepts values between 0.25 and 4.0. The default speed is 1.0. This allows you to optimize audio for accessibility standards, such as slower speeds for dyslexia-friendly listening or faster speeds for quick updates.
Speed adjustments do not affect the character count or billing. Each request is processed sequentially, so longer texts may queue under load.
Response Formats
The API returns synthesized audio in the requested format, ensuring compatibility with your existing video production pipelines. Each file includes an inaudible watermark to identify synthetic speech, which is critical for accessibility compliance and content transparency.
MP3 Output
MP3 is the default format, offering broad compatibility with most video editors and web players. It provides a balanced file size and quality, suitable for standard voice-over workflows.
WAV Output
WAV delivers uncompressed PCM audio, preserving the exact audio fidelity generated by the engine. This format is ideal for archival purposes or when your editing software requires lossless input.
Opus and FLAC Options
Opus provides high-quality compression for efficient storage. FLAC offers lossless compression, reducing file size without sacrificing audio quality, making it suitable for long-form content storage.
Rate Limits
The VoiceCompliance API enforces a standard rate limit of 60 requests per minute per active API key. This limit applies to the /v1/audio/speech endpoint and is designed to ensure stable processing for high-volume video voice-over workflows. Requests are processed sequentially, meaning that under heavy load, synthesis may queue behind other pending tasks.
60 Requests Per Minute
After account activation, the standard quota is 60 requests per minute. This capacity supports consistent audio generation for bulk accessibility projects. Each request consumes one unit of the per-minute allowance, regardless of input text length.
Initial Limit
New accounts begin with a reduced limit of 10 requests per minute. This restriction applies immediately upon key creation and persists until the first top-up is processed. The lower initial rate helps manage server load while verifying account validity.
- Limit: 10 requests/minute
- Duration: Until first top-up
Limit After Top-Up
Upon successful receipt of the minimum $5 USDT top-up via the TRON network, the rate limit automatically increases to the standard 60 requests per minute. No manual support ticket or waiting period is required. The increase occurs within minutes of network confirmation.

Error Handling
The API returns standard HTTP status codes to indicate request outcomes. A 200 OK confirms successful synthesis, while 400 series codes denote client errors such as invalid parameters or exceeded character limits. A 401 Unauthorized response indicates an invalid or expired API key, and 429 Too Many Requests signals that the rate limit has been reached. Server-side issues return 500 or 503 codes.
All errors follow a consistent JSON structure to facilitate programmatic handling. This allows product teams to log failures accurately and trigger appropriate fallback mechanisms in their video production pipelines.
Standard HTTP Codes
The API uses standard HTTP status codes to communicate the result of each request. Successful operations return 200 OK. Client errors, such as invalid voice IDs or malformed JSON, return 4xx codes. Authentication failures return 401, and rate limit exceedances return 429. Server issues result in 5xx codes.
- 200 OK: Request processed successfully.
- 400 Bad Request: Invalid input or parameters.
- 401 Unauthorized: Invalid or missing API key.
- 429 Too Many Requests: Rate limit exceeded.
- 500/503: Internal server or temporary availability issues.
JSON Error Body
Every error response includes a JSON body detailing the specific issue. This structure includes an error code, a human-readable message, and optional parameters indicating which field caused the failure. This consistency allows for automated parsing and logging within CI/CD pipelines.
The response body ensures that debugging is efficient, providing exact error details rather than generic messages. This is critical for high-volume workflows where manual inspection of logs is impractical.
Retry Logic
For transient 5xx errors, implement exponential backoff with jitter. Retry 429 errors after the duration specified in the Retry-After header. Do not retry 4xx errors as they indicate persistent client-side issues. Failed synthesis requests are free, so retrying after a temporary error does not incur additional charges.
- Use exponential backoff for 5xx errors.
- Respect Retry-After header for 429 responses.
- Do not retry 4xx client errors.
- Retry logic should handle network timeouts gracefully.
SDK Compatibility
The API mirrors the OpenAI speech endpoint structure, allowing teams to leverage existing OpenAI SDKs by simply adjusting the base URL. This ensures seamless integration for product teams already familiar with OpenAI’s client libraries.
OpenAI Python SDK
Import the OpenAI Python library and initialize the client by pointing it to our API base URL. Pass your API key via the environment variable or explicitly in the client configuration. The audio.speech.create method works identically to the standard OpenAI implementation.
- Install via
pip install openai - Use
AudioResponseto handle binary audio data - Supports synchronous and asynchronous workflows
OpenAI Node.js SDK
Initialize the OpenAI Node.js client with the custom base URL. Use the audio.speech.create method to generate audio files. The response provides a readable stream or buffer, enabling direct piping to storage or playback systems in Node.js environments.
- Install via
npm install openai - Handle audio data as
Uint8ArrayorBuffer - Compatible with Node.js 18+ environments
Base URL Configuration
Set the base_url parameter to https://api.368g.info/v1 in your SDK client initialization. Ensure the Authorization header includes the Bearer token prefixed with tts-. This configuration directs all requests to the VoiceCompliance infrastructure while maintaining standard SDK compatibility.
Code Examples
The API accepts standard JSON payloads, allowing seamless integration into existing voice-over pipelines. Input text is limited to 4,096 characters per request, requiring splitting for longer scripts.
All responses return binary audio data or structured errors. Use the specified SDKs to handle base URL configuration and authentication headers automatically.
cURL Command
Execute a direct synthesis request using the terminal. Replace YOUR_API_KEY with your active key from /get-api-key/.
The command sends a POST request to the speech endpoint, specifying the model, voice, and output format.
- POST https://api.368g.info/v1/audio/speech
- Content-Type: application/json
- Authorization: Bearer YOUR_API_KEY
Python Script
Use the OpenAI Python SDK, pointing to the custom base URL. Initialize the client with your Bearer token.
Call audio = client.audio.speech.create() with the desired voice and input text. Save the binary response to a file.
Node.js Script
Initialize the OpenAI Node.js SDK with the base URL https://api.368g.info/v1. Pass your API key in the auth header.
Invoke the speech generation method. Handle the stream or buffer to save the audio file locally for your video projects.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. His natural pacing aids comprehension, particularly for accessibility-focused materials where clear, approachable speech is prioritized over formal narration.
23 languages
Any voice speaks any of these languages.
