Frequently asked questions
What is a text reader API?
Q: What is a text reader API? A: It is a cloud text-to-speech (TTS) API. You send text and receive an audio file. The endpoint is POST https://api.play2jj.com/v1/audio/speech with a JSON body. It uses the same request shape as the OpenAI speech endpoint, allowing official OpenAI SDKs to work by changing the base_url to https://api.play2jj.com/v1.
How does the text reader work for language learning?
Our API converts text to speech via a POST request to /v1/audio/speech. You receive an audio file in your chosen format. You can specify the voice, speed, and language to match your curriculum needs. This allows language learning startups to integrate high-volume audio generation directly into their educational tools programmatically, ensuring consistent quality for student practice.
Which languages are supported?
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. Any voice can speak any of these languages. Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, Bengali, and others are not supported.
How many voices are available?
There are eight voice presets available. Female voices include aria, nova, luna, and sage. Male voices include felix, milo, atlas, and orion. These voices are based on licensed recordings from LibriTTS-R and LibriVox public-domain audiobook readings. Each generated file carries an inaudible watermark marking it as synthetic speech.
What audio formats can I use?
Select from mp3, wav, opus, flac, or pcm. Specify your choice in the response_format field of the request body. The API returns the audio file directly in the requested format, ensuring compatibility with your language learning applications and playback systems. Note that each request supports up to 4096 characters of input text.
How much does the text reader cost?
Our API charges $10 per 1 million characters ($0.01 per 1,000 characters), including spaces and punctuation. Failed requests are free. There are no monthly plans, seats, or minimum spend requirements. Your account balance never expires. Note that the $0.50 free trial credit expires after 7 days, while top-up funds remain valid indefinitely.
Are failed requests charged?
No, failed requests are free. If synthesis fails, the charge is returned in full. You only pay for successful audio generation. This ensures you are billed only for usable audio files delivered to your application, reducing risk for high-volume language learning text processing.
Is there a free trial?
Yes, a free trial provides $0.50 of credit, valid for 7 days, when you sign in with Google. This covers about 50,000 characters. Accounts created with email and password start without trial credit. The trial allows testing the API before committing to USDT top-ups for production use.
How do I pay with USDT?
Payments are made via USDT on the TRON network (TRC-20) only. The minimum top-up is $5. Credits are added automatically within minutes after the transfer. There are no cards, PayPal, subscriptions, or invoices with VAT. Balance does not expire, allowing flexible funding for ongoing language learning content generation.
What is the rate limit?
The limit is 60 requests per minute per key. Before the first top-up, the limit is 10 requests per minute. Requests are processed one after another, so under load, a request may wait in a queue. This ensures consistent quality and prevents overload during high-volume language learning text synthesis.
Is my input text used for training?
No, input text is used only to produce the audio and is not used for training. Generated audio is returned in the response and not stored. Request metadata, such as time, characters, and status, is kept for billing purposes. This ensures data privacy for your language learning content.
Can I specify the language explicitly?
Yes, you can set the optional language field to an ISO code. If not set, the language is guessed from the script, defaulting to English for Latin text. For non-English Latin-script text, you should set the language explicitly to ensure accurate synthesis by the chosen voice preset.
How long does synthesis take?
Typical speed is roughly real time on long texts, about 25 characters per second. Short phrases take a few seconds. Requests are processed one after another, so under load, a request may wait in a queue. This ensures consistent processing order and predictable timing for your language learning application.
Do I need to split long texts?
Yes, split texts into parts of up to 4096 characters per request. Join the resulting audio files after synthesis. The API processes each part separately, so splitting ensures you stay within the character limit. This workflow allows handling longer educational content while maintaining consistent audio quality and format.
First request
curl
curl https://api.play2jj.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.mp3API Integration for Language Apps
Integrate the LingoVoice API into your language learning platform using a standard OpenAI-compatible endpoint. Send a POST request to /v1/audio/speech with your JSON payload and receive a synthesized audio file. This structure allows you to use existing OpenAI SDKs by simply adjusting the base_url, ensuring a smooth transition for your engineering team.
Standard OpenAI-compatible endpoint
Use the familiar speech endpoint structure to generate audio. The API accepts standard parameters like model, voice, and input text, making it easy to plug into your existing backend services without building custom integration logic from scratch.
- POST https://api.play2jj.com/v1/audio/speech
- Supports official OpenAI SDKs via base_url adjustment
- Consistent JSON request and response schema
Simple JSON request structure
Define your voice, model, and input text in a clean JSON body. You can specify the response format (mp3, wav, opus, flac, pcm) and speed (0.25–4.0). The language parameter is optional; if omitted, the system guesses the language, though you should set it explicitly for non-English Latin-script text.
- Fields: model, input, voice, response_format, speed, language
- Input text limited to 4096 characters per request
- Audio formats: mp3, wav, opus, flac, pcm
Instant audio delivery
Upon successful synthesis, the audio file is returned directly in the response. Requests are processed sequentially, so under high load, a request may wait in a queue. Typical generation speed is roughly real-time, but short phrases may take a few seconds. Failed requests incur no charge.
- Audio returned immediately in the response body
- Processed sequentially; queueing possible under load
- Failed requests are free
Voice Selection for Learners
LingoVoice offers eight distinct voice presets to enhance language education tools. Female voices include Aria for warm narration, Nova for upbeat content, Luna for soft storytelling, and Sage for calm instruction. Male options feature Felix for energetic promos, Milo for friendly conversation, Atlas for neutral news, and Orion for deep documentary styles.
Clear female narrators for listening practice
Select Aria for clear, warm narration or Luna for soft storytelling. These voices provide distinct tonal qualities suitable for different listening exercises and character differentiation in educational apps.
- Aria: warm, clear narrator
- Luna: soft, storytelling
Neutral male voices for news and dialogue
Use Atlas for neutral, news-style delivery or Milo for friendly, conversational tones. This distinction allows precise matching of audio to content context, ensuring appropriate tone for dialogue or informational segments.
- Atlas: neutral, news
- Milo: friendly, conversational
Distinct tones for character differentiation
Each voice carries a specific character profile, from Orion’s deep documentary style to Felix’s energetic promo tone. This variety supports immersive language learning experiences by distinguishing characters or content types effectively.

Multi-Language Support
Support multilingual content delivery with 23 languages, including Asian and European options. This ensures accurate pronunciation across diverse learner demographics without requiring separate voice models for each region.
23 supported languages including Asian and European
The API supports 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, allowing flexible casting for educational scenarios.
- Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish
Automatic language detection
If the language field is omitted, the system guesses the language from the script. Latin script text defaults to English, ensuring immediate functionality for common use cases without additional configuration.
Manual language code override
For non-English Latin-script text, set the language parameter explicitly to ensure correct pronunciation. This prevents default English inference from affecting accuracy in languages like Dutch or Danish when using standard voices.
Audio Formats for Learning Tools
Audio format selection directly impacts how your language learning application handles playback, storage, and analysis. Our API supports MP3, WAV, OPUS, and FLAC, allowing you to balance file size with fidelity based on your specific technical requirements.
MP3 for mobile app compatibility
MP3 provides broad compatibility across all major mobile platforms and web browsers. This ensures consistent playback behavior for learners without requiring additional codec libraries or plugins within your application.
WAV for high-quality offline analysis
WAV delivers uncompressed audio data, preserving the full waveform generated by the synthesis engine. This format is ideal for offline applications that require raw audio inspection or simple waveform visualization without decompression overhead.
OPUS and FLAC for efficient storage
OPUS offers high-quality audio with efficient compression, suitable for streaming or storage-constrained environments. FLAC provides lossless compression, reducing storage costs compared to WAV while maintaining bit-perfect audio quality for archival purposes.
Scalable Character Pricing
LingoVoice applies a flat rate of $0.01 per 1,000 characters, ensuring predictable costs for language education platforms. Spaces and punctuation count as characters, providing precise billing granularity for your API integration.
This transparent model eliminates hidden fees or tiered surprises. Startups can forecast expenses accurately based on character volume, scaling costs directly in line with user engagement and lesson completion rates.
$10 per million characters
The standard rate is $10 for every million characters processed. This unit-based pricing aligns directly with content volume, making it easy to calculate costs for vocabulary drills, reading passages, or dialogue exercises.
- - $0.01 per 1,000 characters
- - Spaces and punctuation count
- - Predictable unit cost
No minimum spend or monthly fees
There are no monthly subscriptions, seats, or minimum spending requirements. You only pay for the API calls you make, allowing early-stage products to operate with minimal overhead and variable cash flow.
- - No monthly plans
- - No minimum spend
- - No seats
Pay only for successful synthesis
If a synthesis request fails, the charge is returned in full. You are billed only for successfully generated audio files, ensuring you do not pay for errors or timeouts in your application.

Reliable Infrastructure
We ensure consistent delivery through a fixed 60 requests per minute limit per key. This predictable capacity prevents resource exhaustion during peak usage windows, allowing your language learning platform to maintain stable performance without unpredictable throttling.
60 requests per minute rate limit
Each API key supports a maximum of 60 requests per minute. This limit applies uniformly, ensuring fair resource allocation. New accounts start with a lower 10 requests per minute threshold until the first top-up is processed.
Automatic queue processing under load
Requests are processed sequentially. During high-volume periods, synthesis jobs enter a queue and are handled one after another. This ensures order integrity but may introduce latency if the queue is full.
Real-time audio generation
Generation speed is roughly real-time for long texts, averaging approximately 25 characters of text per second of audio. For short phrases, synthesis takes a few seconds regardless of length. This predictable timing aids in managing user expectations during playback.
Data Privacy for Student Records
Input text is transmitted solely to produce the target audio file and is not retained for model training or future use. This ensures that student reading materials remain private and are not repurposed for improving the engine’s voice models.
Generated audio is returned directly in the response and is not stored on our servers. Only request metadata—such as timestamp, character count, and status—are kept for billing accuracy.
Text used only for synthesis
The text you send is processed exclusively to generate the audio response. Once the audio is created, the input text is not archived for training purposes.
No training on input data
Your students' reading texts are not used to train the TTS engine. This separation ensures that proprietary or sensitive educational content remains distinct from the model's learning dataset.
- Input text is used only for synthesis
- Input text is not used for training
Metadata kept for billing only
We retain minimal metadata including request time, character count, and status to verify charges. This data is used strictly for billing transparency and does not include the content of the spoken text.
Developer-Friendly Experience
Manage your API access with a single active key per account. You can issue a new key at any time, and the previous key stops immediately, ensuring tight control over your integration environment.
Single API key management
One key handles all requests. Revoke and rotate instantly without service interruption.
Curl and SDK examples
Use the standard OpenAI SDK by changing the base URL to https://api.play2jj.com/v1. Plain HTTP or curl works identically for POST /v1/audio/speech.
- Compatible with official OpenAI SDKs
- Standard JSON request structure
- No proprietary client required
Clear error codes
Failed synthesis returns full credit. Requests are processed sequentially, so high volume may queue. Check status codes for immediate debugging.

Ideal for EdTech Startups
LingoVoice provides a reliable, character-based TTS API designed for startups building language education tools. Our service handles high-volume text-to-speech generation with consistent quality and transparent billing, ensuring your development team can focus on product features rather than audio infrastructure.
Cost-effective for early-stage products
Startups can begin with a $0.50 free trial (valid for 7 days) when signing in with Google. Pricing is straightforward at $10 per million characters, with no minimum spend or monthly fees. Failed requests are free, so you only pay for successful synthesis.
- Free trial: $0.50 credit for Google sign-ups
- Pay-per-use: $0.01 per 1,000 characters
- No hidden fees or subscription locks
Scales with user growth
The API supports up to 60 requests per minute per key, with automatic queue processing under load. Text is split into parts of up to 4,096 characters and joined seamlessly, allowing you to generate long-form educational content without complex client-side logic.
- 60 requests per minute standard rate limit
- Automatic text splitting for long lessons
- Consistent quality across high-volume generations
No infrastructure management
We handle the synthesis engine, so you do not need to manage servers or audio storage. Input text is used only for synthesis and not for training, while generated audio is returned in the response and not stored by us.
- Managed synthesis engine
- No server maintenance required
- Input text not used for training
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Mastering new languages empowers your startup to expand globally and connect with diverse talent pools effortlessly.
23 languages
Any voice speaks any of these languages.
