Frequently asked questions
What is the Polyglot Cloud TTS API?
It is a cloud text-to-speech API that accepts text and returns an audio file. The endpoint is POST https://api.play15q.com/v1/audio/speech with a JSON body. It uses the same request shape as the OpenAI speech endpoint, allowing you to use official OpenAI SDKs by simply changing the base URL to https://api.play15q.com/v1.
How does the Polyglot Cloud TTS API work?
Send a POST request to the speech endpoint with your text, model, voice, and format. The API processes the request and returns the generated audio file. For texts exceeding 4096 characters, split the input into parts, generate audio for each, and join the files. Requests are processed sequentially, so high volume may cause queue delays.
Is the Polyglot Cloud TTS API free to use?
No, it is a paid service, but new accounts signing in with Google receive a $0.50 free trial valid for 7 days. Email-based accounts start without trial credit. Failed requests are free, with charges returned in full. Otherwise, you pay per character using USDT on the TRON network (TRC-20).
What languages does the Polyglot Cloud TTS API support?
It supports 23 languages: 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. If the language is not set, it is guessed from the script.
How much does the Polyglot Cloud TTS API cost?
The cost is $10 per 1 million characters, or $0.01 per 1,000 characters. Spaces and punctuation count as characters. You can top up your balance with USDT on the TRON network, with a minimum of $5. There are no monthly plans, seats, or minimum spends, and the balance does not expire.
Can I use the Polyglot Cloud TTS API with Python?
Yes, because the request shape matches the OpenAI speech endpoint, you can use the official OpenAI Python SDK. Simply configure the SDK’s base URL to https://api.play15q.com/v1 and provide your API key. This allows you to use standard Python libraries to send text and receive audio files seamlessly.
Does the Polyglot Cloud TTS API support SSML?
No, the Polyglot Cloud TTS API does not support SSML. It also lacks phoneme dictionaries, custom pronunciation lexicons, emotion tags, and custom voice cloning. You control the output by selecting from eight preset voices and adjusting the speed between 0.25 and 4.0.
What is the character limit for the Polyglot Cloud TTS API?
The input text is limited to 4,096 characters per request. For longer texts, you must split the content into parts of up to 4,096 characters each. Generate audio for each part separately and then join the resulting audio files to create the final output.
How does the Polyglot Cloud TTS API compare to AWS Polly?
Unlike AWS Polly, this API offers a simple JSON endpoint compatible with OpenAI SDKs and charges per character rather than per request. It supports 23 languages with eight distinct voice presets derived from licensed recordings. It lacks SSML and custom cloning but provides straightforward integration for indie developers building language-learning apps.
Can I customize voices in the Polyglot Cloud TTS API?
No, you cannot customize voices or upload your own for cloning. The API provides eight presets: Aria, Nova, Luna, Sage, Felix, Milo, Atlas, and Orion. Each voice is based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain audiobook readings. You can only adjust the speaking speed between 0.25 and 4.0.
Does the Polyglot Cloud TTS API store my text?
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, including time, characters, and status, is kept for billing purposes. No other data is retained.
What audio formats does the Polyglot Cloud TTS API output?
The API outputs audio in mp3, wav, opus, flac, or pcm formats. You specify the desired format in the request body using the response_format field. Every generated file carries an inaudible watermark that marks it as synthetic speech.
Is the Polyglot Cloud TTS API suitable for real-time use?
It is suitable for near real-time use, generating roughly 25 characters of text per second of audio. However, requests are processed sequentially, so under load, a request may wait in a queue. It does not support streaming or WebSocket connections for partial audio delivery.
How do I get an API key for the Polyglot Cloud TTS API?
Sign up on the /get-api-key/ page using Google or an email and password. Accounts are per site. Once registered, the page displays your API key, balance, and usage statistics. You can issue a new key at any time, which immediately invalidates the old one.
Definition and Core Function
Polyglot Cloud TTS is a straightforward text-to-speech API designed for developers integrating audio into language-learning applications. It converts written text into audible speech files via a simple HTTP request, allowing developers to generate pronunciation guides or lesson content programmatically.
Cloud-based text-to-speech conversion
The service runs entirely in the cloud, handling speech synthesis on remote servers. This eliminates the need for local processing power or large voice model files on your device. You send text, and receive an audio file back, keeping your infrastructure lightweight.
Input text to audio file output
Input is plain text, limited to 4096 characters per request. The API processes this input and returns a generated audio file. Supported formats include MP3, WAV, OPUS, FLAC, and PCM, ensuring compatibility with various playback environments.
API-driven integration
Integration relies on standard HTTP POST requests to the /v1/audio/speech endpoint. The JSON payload structure mirrors OpenAI’s speech endpoint, enabling immediate use of existing OpenAI SDKs by simply adjusting the base URL. Authentication is handled via a Bearer token.
How It Works for Developers
Developers interact with the Polyglot Cloud TTS API via standard HTTP POST requests to the endpoint https://api.play15q.com/v1/audio/speech. This design allows you to use the official OpenAI SDKs simply by adjusting the base_url configuration, or send raw JSON payloads via any HTTP client.
The API expects a JSON body containing your text, chosen voice, and optional parameters like speed or language. Upon successful processing, the service returns the generated audio file directly in the response, ready for immediate playback or storage in your language-learning application.
HTTP POST requests
Send a POST request to https://api.play15q.com/v1/audio/speech. Include your API key in the Authorization: Bearer tts-... header. The endpoint accepts standard JSON bodies, making integration straightforward for indie developers familiar with RESTful patterns.
JSON payload structure
The request body requires the input text (up to 4096 characters), a model (tts-1 or tts-1-hd), and a voice preset. You may optionally specify response_format, speed (0.25–4.0), and language (ISO code). Texts exceeding 4096 characters must be split into multiple requests and joined on the client side.
input: The text to synthesizemodel: 'tts-1' or 'tts-1-hd'voice: One of 8 available presetsspeed: Float between 0.25 and 4.0language: Optional ISO code (e.g., 'en', 'de')
Audio file delivery
Successful requests return the audio file directly in the response body. Supported formats include mp3, wav, opus, flac, and pcm. Failed requests incur no charge, and the audio is returned in the response and not stored.
Pricing Model Overview
The Polyglot API uses a transparent per-character billing model, charging $10 per million characters ($0.01 per 1,000 characters). Spaces and punctuation count toward the total, providing predictable costs for your language-learning app’s audio generation needs.
Per-character billing
Costs scale linearly with usage. Failed requests incur no charge, ensuring you only pay for successful audio synthesis. This model suits projects with variable traffic, from small pilot apps to high-volume educational platforms.
- Rate: $10 per 1,000,000 characters
Cost efficiency for volume
For developers generating extensive lesson content, the flat rate avoids tiered surprises. With typical speeds of ~25 characters per second, long texts remain cost-effective. No monthly fees or minimum spends apply.
Free trial availability
Google Sign-In users receive $0.50 in trial credit (approx. 50,000 characters) valid for 7 days. Email/password accounts start without trial credit. Top-ups via USDT (TRC-20) are credited automatically within minutes with no expiration.
- Trial: $0.50 for Google sign-ups
Comparison with Other TTS Services
Polyglot API offers a straightforward alternative to established players like AWS Polly and ElevenLabs, focusing on cost-efficiency and multi-lingual support for indie developers. While AWS Polly provides extensive voice options, Polyglot simplifies integration with a familiar OpenAI-style SDK and competitive per-character pricing.
Unlike ElevenLabs, which offers voice cloning and emotional nuance, Polyglot delivers clear, standard voices suitable for educational content at a fraction of the cost. This makes it ideal for language learning apps where volume and budget matter more than hyper-realistic emotion.
Polyglot vs AWS Polly
AWS Polly requires complex routing for multi-language support, whereas Polyglot handles 23 languages with 8 consistent voices. Polyglot’s per-character billing ($0.01/1k chars) is often more transparent than AWS’s tiered pricing structures.
Integration is simpler with Polyglot’s OpenAI-compatible endpoint, allowing developers to swap base URLs without rewriting request logic. AWS Polly’s SDKs are robust but can be heavier for lightweight language-learning apps.
- - Transparent per-character pricing vs. AWS tiered rates
- - 23 languages with 8 consistent voices
- - OpenAI SDK compatibility for easy migration
Polyglot vs ElevenLabs
ElevenLabs offers superior voice cloning and emotional depth, but at a premium cost. Polyglot focuses on clarity and speed, providing 8 presets optimized for educational clarity rather than cinematic realism.
For language learning, clear articulation matters more than emotional variance. Polyglot’s voices are designed for clarity, ensuring learners hear consistent pronunciation across 23 languages without paying for unused emotional features.
- - 8 clear presets vs. ElevenLabs' extensive library
- - Lower cost for high-volume educational content
- - Focus on clarity over emotional nuance
Feature set differences
Polyglot lacks SSML, streaming, and voice cloning. However, it compensates with a simple 4096-character limit per request and fast generation speeds. This simplicity reduces integration complexity for indie developers.
The API supports formats like MP3, WAV, and OPUS, ensuring compatibility with most learning platforms. Unlike services offering complex voice design, Polyglot’s preset voices are consistent and easy to implement.
- - No SSML or streaming support
- - 8 presets vs. hundreds in competitors
- - Simple 4096-char limit per request
Use Cases in Language Learning
For language-learning apps, consistent audio is critical. The Polyglot API generates pronunciation samples directly from text, ensuring learners hear standardized accents without manual recording overhead. This automation scales effortlessly as course content grows.
Developers can dynamically generate audio for any vocabulary word or sentence. By leveraging the API’s multi-language support, apps can offer high-quality listening exercises in multiple languages using a single technical integration.
Pronunciation audio generation
The API converts text into MP3 or WAV files instantly. Developers can generate distinct audio clips for each vocabulary item, ensuring learners hear the correct phonetics. This is essential for apps focusing on accent reduction or phonetic accuracy.
With 23 supported languages, developers can mix languages in a single response format. This allows for side-by-side comparison audio, helping students distinguish between similar sounds across different linguistic groups.
- Supports 23 languages for diverse learner bases
- Generates MP3, WAV, OPUS, FLAC, or PCM
- Up to 4096 characters per request
Audiobook content creation
Long-form content is split into 4096-character chunks and stitched together. This enables the creation of full lesson audio or short stories without manual file management. The consistent voice quality enhances engagement for passive learning.
Preset voices like 'Luna' (soft storytelling) or 'Atlas' (neutral news) provide consistent tone. This uniformity is vital for maintaining immersion in educational narratives without jarring voice changes between sections.
- Automatic audio joining for long texts
- 8 distinct voice presets available
- Realistic synthesis at near real-time speed
Interactive lesson support
API integration allows for on-demand audio generation during user interactions. When a student clicks a word, the app fetches and plays the audio instantly. This interactivity reinforces memory retention through auditory feedback.
Speed adjustment (0.25–4.0x) allows learners to slow down difficult passages or speed up review sessions. This flexibility accommodates different proficiency levels within the same app without requiring multiple audio files.
- Speed control from 0.25x to 4.0x
- Instant audio delivery via REST API
- No manual audio file storage required
Technical Requirements
The API requires a standard internet connection to send text and receive audio files. Requests are processed sequentially, so under heavy load, synthesis may queue. Ensure your infrastructure handles HTTP POST requests to the base URL without local dependencies.
Internet connection needed
Synthesis happens in the cloud, so your app must handle network latency. Long texts are split into 4096-character chunks and joined client-side or via your backend logic.
API key management
Each account holds one active key. Issuing a new key immediately invalidates the old one, so update your configuration atomically. Keys are passed via the Authorization: Bearer header.
SDK compatibility
The endpoint mirrors the OpenAI speech API structure. You can use official OpenAI SDKs by simply changing the base_url to https://api.play15q.com/v1. Plain HTTP requests also work directly.

Limitations and Constraints
The Polyglot Cloud TTS API imposes specific constraints that developers should account for when designing language-learning applications. Understanding these limits ensures smooth integration and predictable performance during high-volume usage periods.
These constraints primarily affect batch processing, concurrency, and delivery methods. Below are the key technical limitations.
Character limit per request
Each API call accepts a maximum of 4096 characters, including spaces and punctuation. For longer texts, developers must split the input into multiple requests and join the resulting audio files client-side.
Rate limits per key
The API allows 60 requests per minute per key. New accounts start with a lower limit of 10 requests per minute until the first top-up is processed.
No streaming support
The service does not support WebSocket or real-time partial audio streaming. Audio files are returned only after full synthesis is complete, which may introduce latency for interactive use cases.
Data Handling and Privacy
The Polyglot Cloud TTS API is designed with privacy in mind for indie developers. Input text is used solely to generate the requested audio file and is not used for model training. Generated audio is returned directly in the API response and is not stored on our servers after delivery. This ensures your lesson content or audiobook scripts remain yours.
For billing and auditing purposes, we retain request metadata, including the timestamp, character count, and status code. This data is kept only as long as necessary to verify charges and resolve billing disputes. We do not store the actual audio files or the input text beyond the immediate generation process.
Input text usage policy
Your input text is processed exclusively to synthesize the audio file. It is not ingested into our training data or used to improve the underlying voice models. Once the audio is generated and delivered, the text is not retained for future analysis or reuse.
This approach ensures that your proprietary lesson content, vocabulary lists, or audiobook scripts remain confidential. You retain full ownership of the text you submit to the API.
- Text is not used for training models
- Text is not stored after audio generation
- Full ownership of input content remains with the user
Audio storage duration
Generated audio files are returned directly in the HTTP response and are not permanently stored on our servers. If you need to keep the audio, you must download and save it to your own infrastructure immediately after the request completes.
This design minimizes our storage footprint and reduces liability, ensuring we do not hold copies of your generated content indefinitely. You control the persistence of your audio assets.
- Audio is not stored on our servers
- Download immediately after API response
- You control audio persistence
Metadata retention
We retain request metadata, including the timestamp, character count, and status code, for billing verification and dispute resolution. This data is kept only as long as necessary to confirm charges and support your account.
The metadata does not include the input text or the audio content, preserving your privacy while ensuring accurate billing records.
No personal identifiers are linked to the text content itself.
- Metadata includes timestamp, characters, status
- Kept for billing verification
- Not stored indefinitely
Voice Selection and Customization
Polyglot Cloud TTS provides eight distinct voice presets designed to suit various narrative styles, from warm narration to energetic promos. Each voice is derived from licensed recordings, ensuring consistent quality across all generated audio files without requiring custom voice cloning.
Preset voice options
The API offers four female voices (Aria, Nova, Luna, Sage) and four male voices (Felix, Milo, Atlas, Orion). These presets cover a range of tones, including calm instruction, bright upbeat delivery, and deep documentary styles, allowing developers to match the emotional context of their language-learning content.
- Female: Aria, Nova, Luna, Sage
- Male: Felix, Milo, Atlas, Orion
Language selection
All eight voices can speak any of the 23 supported languages, including English, Spanish, German, Japanese, and Arabic. While the system guesses the language from the script by default, explicitly setting the language parameter ensures accuracy, especially for Latin-script texts that might otherwise default to English.
- Supports 23 languages
- Any voice speaks any supported language
Speed adjustment
Users can adjust speech speed between 0.25x and 4.0x using the speed parameter. This flexibility allows learners to hear natural-paced pronunciation or slower, clearer enunciations for difficult sounds, enhancing the educational value of the generated audio.
Error Rates and Reliability
Reliability in TTS depends on predictable error handling and clear billing logic. Our API ensures you are not charged for synthesis failures, maintaining trust during high-volume educational content generation.
Common error types
Errors typically stem from exceeding the 4,096 character limit per request or hitting the 60 requests-per-minute rate limit. Network timeouts may occur during peak load, requiring your client to implement retry logic with exponential backoff.
Refund policy for failures
If synthesis fails, the charge is returned in full. You only pay for successfully generated audio files, ensuring cost efficiency even when minor technical issues disrupt a batch process.
Service uptime notes
We do not offer an SLA or uptime guarantee. Requests are processed sequentially, so under heavy load, a request may wait in a queue before generation begins.
Integration Complexity
Integration is straightforward because the API mirrors the OpenAI speech endpoint structure. Developers can use existing OpenAI SDKs by simply changing the base_url to https://api.play15q.com/v1. A standard POST request with a JSON body containing the text, voice, and format delivers audio files directly, requiring no complex middleware or custom parsing logic.
Ease of implementation
The request shape matches the OpenAI standard, allowing immediate use of familiar libraries. You only need to provide your API key in the Authorization header and specify parameters like model, input, and voice. This reduces boilerplate code significantly for indie developers building language-learning apps.
Documentation quality
Clear endpoint definitions are provided at /docs/, detailing required fields like response_format and speed ranges (0.25–4.0). The documentation includes ready-to-use curl examples and explains character limits (4096 per request) and rate limits (60 requests/minute), ensuring developers can predict integration behavior accurately.
Support
There is no live chat or phone support; assistance is provided via email only. Since the API uses standard HTTP methods and JSON, most integration questions can be resolved by reviewing the endpoint specifications or testing with the free trial credit.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Language learning empowers indie developers to connect deeply with global audiences without heavy marketing budgets.
23 languages
Any voice speaks any of these languages.
