Frequently asked questions
What is the Bet44 TTS API?
It is a cloud text-to-speech API where you send text via POST and receive an audio file. The endpoint is https://api.bet44.info/v1/audio/speech. It uses a JSON body similar to the OpenAI speech endpoint, allowing developers to integrate synthetic speech generation into their applications with predictable character-based pricing and standard HTTP requests.
Is the free trial really free?
Yes, signing in with Google provides $0.50 of credit, equivalent to about 50,000 characters, valid for 7 days. This allows immediate integration testing without payment. Accounts created with email and password do not receive this trial credit. The trial is a one-time benefit per account to verify the API works for your specific use case.
How do I get my API key?
Navigate to the /get-api-key/ page and sign up using Google or an email and password. The page displays your active API key, balance, and usage statistics. You can issue a new key at any time, which immediately invalidates the previous one. The key is used in the Authorization header as a Bearer token.
Which programming languages are supported?
Since the API uses standard HTTP requests, it supports any programming language capable of sending JSON payloads. You can use the official OpenAI SDKs by changing the base_url to https://api.bet44.info/v1. Alternatively, you can use plain HTTP libraries in any language that handles RESTful API calls.
Can I use the OpenAI SDK?
Yes, the request shape matches the OpenAI speech endpoint. You can use the official OpenAI SDKs by configuring the base_url to https://api.bet44.info/v1 and providing your site-specific key. This allows you to leverage existing SDK code with minimal changes, ensuring a smooth integration process for developers already familiar with the OpenAI ecosystem.
What happens if a request fails?
If synthesis fails, the charge is returned in full. Failed requests are free, ensuring you only pay for successful audio generation. This applies to any request that does not complete successfully, protecting your balance from errors during the text-to-speech conversion process. You can retry the request without additional cost if the issue is temporary.
How are characters counted?
Characters are counted based on the input text length, including spaces and punctuation. The price is $10 per 1 million characters, or $0.01 per 1,000 characters. Each request can contain up to 4,096 characters. Longer texts must be split into parts and joined, with each part counted separately for accurate billing and processing.
Is there a minimum spend?
No, there is no minimum spend for using the API. You pay only for the characters you generate. The minimum top-up amount is $5 when adding funds via USDT on the TRON network. Balances do not expire, and there are no monthly plans or subscription fees, allowing flexible usage based on your project needs.
Do you support streaming audio?
No, the API does not support streaming, WebSocket, or real-time partial audio. You send a POST request and receive the complete audio file in the response. This deterministic approach ensures predictable costs and simple integration, as you do not need to handle streaming protocols or partial data chunks during the synthesis process.
Can I clone my own voice?
No, custom voice cloning is not offered. You can choose from eight preset voices based on licensed recordings. There is no option to upload your own voice or design custom voices. The generated audio files carry an inaudible watermark to mark them as synthetic speech, distinguishing them from human recordings without requiring cloning technology.
What voices are available?
There are eight presets: Aria, Nova, Luna, Sage (female) and Felix, Milo, Atlas, Orion (male). Each voice has a distinct tone, from warm narrators to deep documentary styles. All voices can speak any of the 23 supported languages. You select the voice by name in the request body to generate the desired audio output.
How many languages does it support?
It supports 23 languages, including English, Spanish, German, French, Japanese, and Chinese. Any voice can speak any of these languages. If the language is not specified, it is guessed from the script, defaulting to English for Latin text. Unsupported languages include Vietnamese, Indonesian, Thai, Ukrainian, and Romanian, among others not listed in the supported set.
What is the input limit per request?
Each request can contain up to 4,096 characters of text. For longer texts, you must split the input into parts of this size and join the resulting audio files. This limit ensures consistent processing times and predictable billing. Requests are processed sequentially, so under load, longer texts may experience slight delays in queue processing.
How is the API secured?
Security is handled via Bearer token authentication using your unique API key. You pass the key in the Authorization header with each request. Each account has one active key, which can be revoked and replaced instantly. Input text is used only for generation and not for training, ensuring your data remains private during processing.
First request
curl
curl https://api.bet44.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.mp3Developer-First Infrastructure
Our infrastructure delivers a deterministic, character-accurate text-to-speech API tailored for developers who need predictable costs without enterprise overhead. The primary endpoint is POST https://api.bet44.info/v1/audio/speech, accepting standard JSON bodies that mirror the OpenAI speech endpoint structure. This design allows immediate integration using existing OpenAI SDKs by simply adjusting the base_url configuration, eliminating the need for custom client libraries or complex parsing logic.
OpenAI-Compatible Endpoint
The endpoint POST https://api.bet44.info/v1/audio/speech handles all synthesis requests. It accepts the same request shape as the OpenAI speech endpoint, ensuring that developers can switch providers with minimal code changes. By updating the base_url to https://api.bet44.info/v1 and supplying your tts-... API key, standard OpenAI SDKs function correctly without additional adaptation.
- Standard POST method for audio generation
- Compatible with official OpenAI SDKs via base_url override
- Supports model parameters:
tts-1andtts-1-hd
Simple POST Requests
Requests require a JSON body containing model, input (text up to 4096 characters), voice, and response_format. Authentication is handled via the Authorization: Bearer tts-... header. The system processes requests sequentially, ensuring deterministic output. For texts exceeding the character limit, developers must split the input into parts and join the resulting audio files.
- Maximum input length: 4096 characters per request
- Speed parameter range: 0.25 to 4.0
- Optional
languagefield for ISO code override
JSON Response Format
Upon successful synthesis, the API returns the audio file directly in the response body. The format corresponds to the requested response_format (mp3, wav, opus, flac, or pcm). Failed requests return an error JSON object and incur no charge, ensuring that billing reflects only successful character consumption. Metadata such as request time and character count is retained for billing purposes.
Immediate Testing Capability
The platform provides an immediate testing environment for developers seeking to validate integration without financial commitment. Upon signing in via Google, accounts are automatically issued $0.50 of trial credit, equivalent to approximately 50,000 characters. This allows for rapid verification of voice quality, language handling, and endpoint compatibility before any payment details are entered.
This structure eliminates administrative friction, enabling instant access to the API for prototype development. The credit is valid for seven days, providing a sufficient window to complete initial integration tests and confirm that the deterministic, character-accurate billing model meets project requirements.
$0.50 Free Trial Credit
New accounts registered with Google authentication receive an automatic $0.50 credit. This amount covers roughly 50,000 characters of text-to-speech synthesis, allowing developers to generate substantial audio samples to evaluate voice clarity and processing speed.
The credit is applied immediately upon account creation, ensuring no delay between sign-up and the ability to make API requests. This enables immediate validation of the endpoint behavior and response formats without waiting for manual approval or payment verification.
- Equivalent to 50,000 characters
- Automatically applied to Google-signup accounts
- Sufficient for comprehensive voice testing
No Credit Card Required
The trial credit is issued without requiring a payment method on file. Developers can begin making requests immediately after authentication, removing the barrier of payment entry during the initial discovery phase. This is particularly useful for indie developers who prefer to verify technical fit before committing to the USDT top-up process.
Billing only begins when the trial credit is exhausted or when a manual top-up is performed. The absence of a credit card requirement reduces the friction of initial adoption, allowing teams to test the API's stability and character-accurate pricing without risk.
7-Day Trial Window
The $0.50 trial credit remains valid for seven days from the time of account creation. This duration provides ample time to complete integration tests, generate sample audio, and evaluate the API's performance under typical load conditions. If the seven-day period expires before the credit is used, it is forfeited.
This window is designed to support standard development cycles for small projects. Developers can use this time to decide whether the deterministic pricing and voice presets meet their needs before proceeding to the TRC-20 USDT top-up process for ongoing usage.
High-Quality Voice Selection
Vocalift offers eight distinct voice presets designed for specific use cases, ranging from warm narrators to neutral news readers. Each voice is engineered to deliver clear, consistent audio output suitable for indie applications without requiring complex configuration.
8 Distinct Voice Presets
The platform provides four female voices—Aria, Nova, Luna, and Sage—and four male voices—Felix, Milo, Atlas, and Orion. These presets cover diverse tones, including energetic promos, calm instructional guides, and deep documentary styles, allowing developers to match audio to their content's mood.
- 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
Licensed Source Material
All voice models are derived from licensed recordings, ensuring legal compliance for commercial use. The underlying audio data originates from LibriTTS-R (CC BY 4.0), which consists of public-domain audiobook readings from LibriVox. This foundation guarantees high-quality synthetic speech without additional licensing overhead.
Inaudible Watermarking
Every generated audio file includes an inaudible watermark that explicitly marks the content as synthetic speech. This feature aids in content identification and transparency, distinguishing Vocalift’s output from human-recorded audio without affecting the listener's experience or audio quality.
Broad Language Support
The API supports 23 languages, including English, Spanish, German, Japanese, and Arabic. Any voice preset can speak any supported language, ensuring consistent audio quality across diverse linguistic requirements without needing separate voice models for each region.
23 Supported Languages
Supported languages are Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish. Languages such as Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, and Bengali are not supported.
Automatic Language Detection
When the language field is omitted, the system guesses the language from the script. Latin-script text defaults to English. This behavior simplifies initial integration but may produce incorrect accents for non-English Latin text.
Manual Language Override
For non-English Latin-script text, explicitly set the language parameter to the ISO code. This ensures the correct phonetic pronunciation and prevents the system from defaulting to English, which may result in unnatural speech patterns.
Predictable Pricing Model
Our pricing model eliminates financial uncertainty by charging strictly on a per-character basis. You pay exactly what you consume, with no hidden fees or minimum spend requirements. This structure ensures that costs scale linearly with your usage, providing precise predictability for indie developers managing tight budgets.
The absence of monthly subscriptions or seats means you are never locked into recurring costs for unused capacity. Whether you generate 1,000 characters or 100 million, the rate remains constant. Failed requests are refunded in full, ensuring you only pay for successful synthesis results.
Per-Character Billing
Every character in your input text, including spaces and punctuation, is counted toward your total. The standard rate is $0.01 per 1,000 characters. This means 1 million characters cost exactly $10.00. This clarity allows you to calculate exact project costs before making a single API call.
- $0.01 per 1,000 characters
- Spaces and punctuation count
- No minimum spend required
No Monthly Subscriptions
We do not offer monthly plans, seats, or tiered access levels. Your account balance is drawn down as you use the API, and the balance does not expire. You can top up whenever necessary without waiting for a billing cycle or managing recurring invoices.
Transparent Rate Card
The cost is uniform across all models (tts-1 and tts-1-hd) and formats. There are no premium charges for HD quality or specific voices. The only variable is the character count and request success status, making the total cost easy to forecast with high accuracy.

Robust Audio Formats
The API supports MP3, WAV, Opus, FLAC, and PCM formats. Select the appropriate format based on your application’s bandwidth constraints and quality requirements.
MP3 and WAV Support
MP3 offers efficient compression suitable for general distribution, while WAV provides lossless audio.
FLAC and Opus Options
For archival purposes or advanced compression needs, FLAC delivers lossless quality. Opus provides audio quality at low bitrates.
PCM Raw Output
PCM delivers raw, uncompressed digital audio samples.
Technical Specifications
Technical constraints are defined by character limits, speed multipliers, and processing latency. Understanding these parameters ensures accurate request shaping and predictable delivery times for your integration.
The API enforces a strict 4096-character limit per request. Speed ranges from 0.25x to 4.0x. Processing occurs sequentially, meaning high load may introduce queue delays.
4096 Character Limit
Each POST request accepts a maximum of 4096 characters in the input field. Spaces and punctuation count toward this total. For texts exceeding this limit, split the content into multiple requests and join the resulting audio files.
Speed Control Range
The speed parameter accepts values between 0.25 and 4.0. This allows precise control over the audio duration without altering pitch. Default speed is 1.0.
Billing remains strictly per character processed.
- Minimum speed: 0.25x
Real-Time Processing
Generation typically takes about 25 characters of text per second of audio produced. Short phrases may take a few seconds due to initialization overhead. Requests are processed sequentially, so concurrent requests may wait in a queue under load.
Integration Workflow
Authentication relies on a single active bearer token issued per account. Sending a new key instantly invalidates the previous one, ensuring clear audit trails. Pass the token via the Authorization header in every request to the https://api.bet44.info/v1 base URL.
API Key Authentication
Include your key in the Authorization: Bearer tts-... header. Each account holds one active key at a time; issuing a new one immediately deactivates the old one. This simplifies credential management for indie developers without requiring complex rotation logic.
cURL Quick Start
Use a standard POST request to /v1/audio/speech. The endpoint accepts JSON bodies with model, input, voice, and response_format. Receive the generated audio file directly in the response body. This method requires no external libraries, making it ideal for quick validation of voice quality and latency.
curl -X POST https://api.bet44.info/v1/audio/speech-H "Authorization: Bearer YOUR_KEY"-H "Content-Type: application/json"- `-d '{"model": "tts-1", "input": "Hello", "voice": "nova", "response_format": "mp3"}'`
SDK Compatibility
The endpoint mirrors the OpenAI speech schema, allowing direct use of official OpenAI SDKs by changing the base_url to https://api.bet44.info/v1. No custom client library of our own is required. You can integrate the API into existing Python, Node.js, or Go projects with minimal code changes.
Data Privacy and Usage
We process input text exclusively to generate the requested audio file. Your source content is not used for model training or improved in future versions, ensuring your intellectual property remains private and distinct from our dataset.
No Training Data Usage
Generated audio is returned directly in the API response and is not stored on our servers after delivery. Only the request metadata—specifically the timestamp, character count, and processing status—is retained to facilitate accurate billing and usage tracking.
- Input text is processed solely for synthesis.
- No usage for model training or improvement.
- Audio files are not permanently stored.
- Metadata kept for billing accuracy.
Transient Audio Storage
Audio files are delivered via the response stream and are not archived by our platform. This ensures that your generated voice assets remain under your control without unnecessary data retention on our infrastructure.
Metadata Retention Policy
We retain request metadata including time, character count, and status codes. This limited data set is used strictly for billing verification and usage monitoring, providing a transparent audit trail for your API consumption.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Each voice is a deterministic preset, meaning the output remains consistent across identical inputs.
23 languages
Any voice speaks any of these languages.
