Frequently asked questions
How does Japanese text to voice work for dyslexic users?
Our API converts Japanese text into natural speech, aiding users with dyslexia by providing clear audio alternatives. You can select voices like aria or nova to suit your needs. The system handles up to 4096 characters per request, ensuring consistent output. This helps product teams build accessible digital experiences by turning text into natural speech via a simple API.
Is the API compatible with standard screen readers?
Yes, our API returns standard audio files (mp3, wav, opus, flac, pcm) that are fully compatible with standard screen readers. Since we do not offer streaming or WebSocket support, the entire audio file is generated and returned. This ensures reliable playback in assistive technologies without requiring custom integration logic for partial audio chunks.
Can I use the same voices for English and Japanese?
Yes, any of our 8 voice presets can speak Japanese. The system supports 23 languages, including Japanese, and any voice can speak any of these languages. If the language is not explicitly set, it is guessed from the script, but you can also specify it directly for consistent results across different languages.
How are characters counted in Japanese text?
Characters are counted exactly as they appear in your input, including spaces and punctuation. The price is $0.01 per 1,000 characters, so Japanese text is billed based on the total character count. This includes all Hiragana, Katakana, and Kanji characters, ensuring transparent and predictable costs for your accessibility features.
Is there a limit on text length per request?
Yes, each request supports up to 4096 characters of input text. For longer texts, you should split the content into parts of this size and join the resulting audio files. This ensures optimal processing speed and reliability, allowing your app to handle long-form content without exceeding the single-request limit.
Do you support SSML for accessibility cues?
No, we do not support SSML (Speech Synthesis Markup Language). Our API uses a simple JSON body with fields like voice, speed, and input. If you need specific pacing or pauses, you can adjust the speed parameter between 0.25 and 4.0, or manage timing in your application logic.
Can I clone a specific voice for brand consistency?
No, we do not offer custom voice cloning for customers. You cannot upload your own voice or design a new one. You must choose from our 8 licensed presets based on recordings from LibriTTS-R and LibriVox. This ensures consistent quality but does not allow for unique brand-specific voice cloning.
Is the audio generated in real-time or pre-rendered?
Audio is generated on-demand via our cloud API, not pre-rendered. While typical speed is roughly real-time for long texts, short phrases may take a few seconds. Requests are processed sequentially, so under load, a request may wait in a queue. We do not offer real-time streaming or WebSocket endpoints.
Do you store the input text after generation?
No, input text is used only to produce the audio and is not used for training. The generated audio is returned in the response and not stored. We keep request metadata, such as time, characters, and status, strictly for billing purposes. Your content remains private and is not retained for other uses.
Can I use the API for educational apps?
Yes, our API is ideal for educational apps, especially for users with dyslexia or visual impairments. You can use voices like sage for instructional content or luna for storytelling. The simple REST API allows easy integration into web or mobile apps, providing accessible audio content for students and learners.
How fast is the API response for long texts?
Typical speed is roughly real-time for long texts, generating about 25 characters of text per second of audio. Short phrases may take a few seconds to process. Since requests are processed one after another, high load may cause queuing delays. This ensures consistent quality without the complexity of streaming endpoints.
Are there any hidden fees for accessibility features?
No, there are no hidden fees for accessibility features. The price is $0.01 per 1,000 characters, with no monthly plans or seats. Failed requests are free, and the charge is returned in full. You pay only for the characters generated, making costs predictable for your accessibility projects.
What happens if a request fails?
If synthesis fails, the charge is returned in full. You are not billed for failed requests, ensuring you only pay for successful audio generation. This protects your budget and allows you to retry requests without worrying about additional costs for errors during the text-to-speech conversion process.
Can I switch voices dynamically in my app?
Yes, you can switch voices dynamically by sending a new request with a different voice parameter. Our API supports 8 presets, including aria, nova, luna, and sage. You can change the voice for each request, allowing your app to offer varied audio experiences without needing to manage multiple voice models.
First request
curl
curl https://api.playmx52.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.mp3Why Accessibility Matters in TTS
For users with dyslexia, reading dense text can be exhausting. Synthetic speech converts written content into natural audio, allowing users to consume information at their own pace without visual strain. This approach reduces cognitive load and makes digital products more inclusive.
Supporting Dyslexic Users
Text-to-speech technology helps bypass reading difficulties by delivering clear, consistent narration. Users can listen to articles, instructions, or product descriptions while reading along, reinforcing comprehension and reducing fatigue during extended usage sessions.
- Reduces visual strain for extended reading
- Supports multi-modal learning styles
- Improves content retention for dyslexic readers
Assisting Visual Impairments
For users with visual impairments, audio is often the primary interface. High-quality TTS ensures that screen readers are supplemented by natural-sounding voices, making navigation and content consumption smoother and less robotic.
- Enhances screen reader alternatives
- Provides natural-sounding navigation cues
- Supports various degrees of vision loss
Compliance Made Simple
Integrating accessible TTS helps meet standards like WCAG without complex custom development. By providing clear audio alternatives for text content, product teams demonstrate a commitment to inclusive design and legal compliance.
- Simplifies WCAG adherence
- Demonstrates inclusive design commitment
- Reduces legal risk in digital products
Japanese Text to Voice Quality
We prioritize accurate Japanese pronunciation and natural rhythm to ensure text-to-speech output feels human and accessible. Our API handles complex kanji and kana combinations with precision, reducing cognitive load for users with dyslexia.
Natural Pronunciation
Japanese text is processed with attention to pitch accent and natural pauses. This clarity helps users with visual impairments navigate content more easily, as the speech flows logically without robotic stumbles.
Context-Aware Generation
The model understands linguistic context to adjust intonation appropriately. Whether reading a simple sentence or a complex paragraph, the output maintains a consistent, readable tone that aids comprehension.
Consistent Audio Output
Every generation delivers uniform quality across different texts. This reliability means your product teams can trust the audio experience remains stable, supporting long-form reading sessions without fatigue.
Developer-Friendly API
Integrate our TTS API using familiar tools. Our endpoint mirrors the OpenAI speech schema, allowing you to leverage existing OpenAI SDKs by simply updating the base URL to https://api.playmx52.com/v1. This compatibility reduces boilerplate code and speeds up development for teams already using these libraries.
OpenAI SDK Compatible
Use standard libraries like the official OpenAI Python or JavaScript SDKs. Pass your API key via the Authorization: Bearer header, and the SDK handles serialization automatically. This means you can switch providers or test our engine without rewriting your integration logic.
Simple JSON Requests
Send a POST request to /v1/audio/speech with a straightforward JSON body. Specify the model, input text, voice, and response_format. The API returns the audio file in the response body, making it easy to save to your application’s storage layer.
- Supports
tts-1andtts-1-hdmodels - Handles up to 4096 characters per request
- Returns audio in mp3, wav, opus, flac, or pcm
No Complex Integrations
No custom SDKs or proprietary wrappers to learn. The API works with standard HTTP clients, curl, or any language that can send JSON. You control the flow: split long texts into chunks, send them sequentially, and concatenate the resulting audio files for a seamless listening experience for your users.
Voice Selection for Clarity
Choosing the right voice ensures your digital content is consumed clearly by users with dyslexia or visual impairments. Our eight preset voices are designed for distinct listening contexts, allowing you to match the tone of your application precisely.
Clear Narrator Voices
Aria offers a warm, clear narration style ideal for long-form reading. Luna provides a soft, storytelling cadence that reduces listener fatigue during extended sessions.
- Aria: Warm, clear narrator
- Luna: Soft, storytelling voice
Conversational Tones
For interactive elements or help texts, Milo delivers a friendly, conversational sound. Nova adds a bright, upbeat energy suitable for onboarding flows or notifications where engagement matters.
- Milo: Friendly, conversational
- Nova: Bright, upbeat
Neutral News Style
Atlas provides a neutral, news-style delivery that remains consistent and easy to parse. Orion offers a deep, documentary tone for serious or educational content requiring gravity.
- Atlas: Neutral, news style
- Orion: Deep, documentary
Supported Languages
Our API handles Japanese text with the same precision as Latin scripts, ensuring consistent natural speech for diverse user bases. This flexibility allows product teams to serve global audiences without managing separate language models or voice assets.
By supporting multiple scripts within a single unified endpoint, we simplify integration for teams building inclusive digital experiences that must accommodate varying literacy levels and visual impairments across different regions.
Japanese Script Support
The engine processes Japanese characters seamlessly, delivering clear, natural-sounding speech without requiring complex preprocessing. This ensures that users with dyslexia or visual impairments receive consistent audio quality regardless of the input script.
No additional configuration is needed; the API automatically detects and optimizes pronunciation rules for Japanese text, maintaining high clarity in generated audio files.
- Consistent output quality across all Japanese text inputs
Latin Script Defaulting
When no language is specified, Latin script text defaults to English. This simplifies integration for common use cases while remaining predictable for developers who do not explicitly set the language parameter.
For non-English Latin-script text, we recommend explicitly setting the language parameter to ensure accurate pronunciation and intonation, avoiding any ambiguity in speech generation.
23 Languages Available
LuminaRead supports 23 languages, allowing any voice to speak any supported language. This flexibility means you can maintain a consistent voice identity across your product while serving users in different linguistic contexts.
Supported languages include Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish. Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, and Bengali are not yet supported.
- Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish
Pricing for Teams
LuminaRead charges $0.01 per 1,000 characters, including spaces and punctuation. This flat rate means your costs scale directly with usage, making it easy to predict expenses for newsletters, e-learning modules, or onboarding flows without hidden fees.
Pay Per Character
You pay only for what you generate. Failed requests are free, so you aren’t charged for errors. With a starting balance of $0.50 from the Google sign-up trial, you can test integration with about 50,000 characters before committing funds.
- $0.01 per 1,000 characters
- Failed requests are free
- Start with $0.50 trial credit
No Monthly Fees
There are no subscriptions, seats, or minimum spend requirements. You keep your balance until it is used, and it never expires. Add funds only when needed, keeping your overhead low and flexible.
Transparent Costs
Top-ups are simple via USDT (TRC-20), credited automatically within minutes. No invoices, no VAT, no surprise charges. Your dashboard shows exactly how many characters remain, so you can plan your accessibility features with confidence.
Audio Formats for Apps
Choose the right audio format to balance playback performance with file size. Our API supports multiple output types, allowing you to tailor the delivery method to your specific application requirements and user device capabilities.
MP3 for Web
MP3 is the standard for web delivery due to its small file size and broad compatibility. It ensures fast loading times for users on slower connections, making it ideal for playback in mobile apps and browsers.
- Small file size reduces bandwidth usage
- Universal support across all modern browsers
WAV for Quality
WAV provides uncompressed audio. Use this format when you need lossless quality.
- Lossless audio quality
- Ideal for high-fidelity playback
FLAC for Archiving
FLAC offers lossless compression, preserving full audio detail while reducing storage space compared to WAV. It is perfect for archiving generated speech for future reference or offline use.
- Lossless compression
- Efficient storage for long-term archives
Data Privacy
We process input text strictly to generate the audio file returned in the response. Your source content is not used for model training or improvement, ensuring that proprietary product copy or user data remains yours.
No Training on Input
Input text is used solely for the immediate production of the audio file. We do not retain your text for retraining our neural networks, keeping your data distinct from our training corpus.
No Audio Storage
Generated audio files are returned directly in the API response and are not stored on our servers after delivery. Only request metadata—time, character count, and status—is kept for accurate billing.
Secure Generation
Audio synthesis occurs in our cloud environment. Each file carries an inaudible watermark marking it as synthetic speech, providing clear attribution without storing the content long-term. Request metadata is retained only for billing purposes.
Get Started Today
Start by visiting the /get-api-key/ page and signing up with your Google account. This creates your independent account on playmx52.com, giving you immediate access to your API key and balance dashboard.
Sign Up with Google
Registration takes seconds. Once signed in, you see your active key, usage stats, and a ready-to-use curl example on your dashboard.
- Instant account creation
- Single key management
- No credit card required
Free Trial Credit
New Google sign-ups receive $0.50 in trial credit, valid for 7 days. This covers roughly 50,000 characters, allowing you to test Japanese TTS quality before committing.
- $0.50 initial credit
- 7-day validity
- No payment info needed
Quick Integration
Use the official OpenAI SDK by pointing it to https://api.playmx52.com/v1. Send a JSON body with your text and desired voice to receive audio instantly.
- OpenAI SDK compatible
- Simple POST requests
- Immediate audio response
Listen to the 8 voices
Each sample below was generated by this API from the same sentence.
23 languages
Any voice speaks any of these languages.