Frequently asked questions
What is the difference between Web Speech API and a cloud TTS API?
The Web Speech API runs locally in your browser, while our cloud TTS API generates audio via a simple HTTP POST request. You send text and receive an audio file, allowing consistent quality across devices without relying on the user’s browser engine or network conditions for synthesis.
Does the cloud API support the same voices as Chrome?
No. We offer eight specific presets like Aria, Nova, and Felix, based on licensed recordings. Chrome uses its own system voices. Our voices are distinct, consistent, and not tied to any single operating system or browser, ensuring uniform output for your web application regardless of the user’s device.
How many languages are supported in the cloud API?
We support 23 languages, including English, Spanish, German, French, Japanese, and Chinese. Any voice can speak any supported language. If you do not specify a language, it is guessed from the script, though you should set it explicitly for non-English Latin text to ensure accuracy.
Can I use the OpenAI SDK with this API?
Yes. Our endpoint matches the OpenAI speech endpoint shape. You can use the official OpenAI SDK by changing the base_url to https://api.playv57.com/v1 and providing your site’s key. This allows you to integrate our service into existing OpenAI-compatible codebases with minimal changes.
What happens if a synthesis request fails?
If synthesis fails, the charge is returned in full. You are not billed for failed requests. This ensures you only pay for successfully generated audio files, protecting your budget from errors during the generation process or temporary service issues.
Is my input text stored after generation?
No. Your input text is used only to produce the audio and is not stored for training. The generated audio is returned in the response and not stored. We keep request metadata like time and status for billing purposes, but not the actual content.
Do I need to pay for a monthly subscription?
No. There are no monthly plans, seats, or minimum spends. You pay only for what you use at $10 per 1 million characters. You can top up your balance with USDT, and your balance does not expire, giving you full control over your spending.
What is the character limit per request?
Each request supports up to 4096 characters. For longer texts, you should split them into parts of this size and join the resulting audio files. This ensures efficient processing and maintains consistent audio quality across your entire document or script.
Can I change my API key if it gets compromised?
Yes. You can issue a new key at any time from your dashboard. When you generate a new key, the old one stops working immediately. This allows you to quickly secure your account and revoke access to any potentially compromised credentials.
Are there any hidden fees for using the API?
No. The price is $0.01 per 1,000 characters, including spaces and punctuation. Failed requests are free. There are no subscription fees, per-seat costs, or hidden charges. You pay exactly for the characters you successfully generate, with transparent billing.
How long is the free trial valid?
The 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 do not receive this trial credit, so you start with a zero balance.
Can I use the generated audio commercially?
Yes. The voices are based on licensed recordings from LibriTTS-R (CC BY 4.0), from LibriVox public-domain audiobook readings. You can use the generated audio in your projects. Every generated file carries an inaudible watermark that marks it as synthetic speech.
What payment methods are accepted?
We accept USDT on the TRON network (TRC-20) only. There are no credit cards, PayPal, or invoices. Top-ups are credited automatically within minutes after transfer. This method allows for quick, low-cost deposits without the friction of traditional payment processors.
Is there a limit to how many characters I can process daily?
There is no daily character limit. The only constraint is the rate limit of 60 requests per minute per key. You can process as many characters as you want, provided you stay within the per-minute request quota and have sufficient balance.
First request
curl
curl https://api.playv57.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.mp3The Browser Limitation
Browser-based speech synthesis relies on the operating system’s built-in voices, leading to inconsistent audio quality depending on the user’s device and browser. While Chrome offers a robust set of voices, Firefox and Safari often fall back to a smaller, less polished library, creating a fragmented user experience.
Inconsistent voices across browsers
When you deploy a web app, you cannot guarantee which voice engine will render the speech. A voice that sounds natural in Chrome may sound robotic or absent in Safari. This inconsistency forces developers to build complex fallback logic or accept that their voice feature will behave differently for every user.
- Chrome uses Google’s neural voices; Safari uses Apple’s voices
- Voice availability changes with OS updates
- No standard for voice personality across platforms
No control over audio format
The Web Speech API outputs audio directly to the speaker or via a basic Blob. You cannot easily capture the audio stream for storage, further processing, or playback in a custom <audio> element with precise control. This limits integration with other media workflows or archival needs.
Limited language support in Chrome
While Chrome supports many languages, the selection is not universal. Some languages available in cloud APIs are missing from the browser’s engine, or vice versa. Relying on the browser means you are capped by Google’s voice library, which does not always align with your app’s global audience needs.
Why Choose Cloud TTS?
The browser’s Web Speech API is convenient but inconsistent. Voice quality, latency, and available languages vary significantly between Chrome, Firefox, and Safari. You cannot reliably control the audio output format, often leaving you with browser-default containers that may not suit your web app’s requirements.
Consistent audio quality everywhere
Cloud TTS eliminates browser variance. The same API call produces identical audio quality whether the user is on Windows, macOS, or Linux. Your voice feature behaves predictably across all environments, removing the guesswork of client-side synthesis.
Full format control (MP3, WAV, OPUS, FLAC, PCM)
Specify the exact output format you need. Choose MP3 for small file sizes, WAV for uncompressed quality, or OPUS and FLAC for efficient streaming. You can also request raw PCM data if your pipeline requires it. No more dealing with unpredictable browser defaults.
Programmatic generation via API
Generate speech entirely through code. Send a JSON POST request and receive an audio file directly in your response. This allows you to integrate voice synthesis into your backend workflows, automate content creation, or build custom voice features without relying on JavaScript’s limited SpeechSynthesis API.
Our API Features
Our API delivers 8 distinct voice presets, ranging from the warm narration of Aria to the deep documentary tone of Orion. Each voice is based on licensed recordings, ensuring consistent quality without the variability of browser engines.
Support covers 23 languages, including Arabic, Chinese, and Swahili, with any voice capable of speaking any supported language. If the language is not set, it is guessed from the script (defaulting to English for Latin text), so you should set it explicitly for non-English Latin text. This eliminates the need for multiple browser instances or platform-specific fallbacks, providing uniform output across all web apps.
8 distinct voice presets
We offer eight curated voices, including Luna for soft storytelling and Felix for energetic promos. These presets are based on licensed recordings from LibriTTS-R, ensuring a consistent, professional tone that does not rely on the user's specific browser or operating system.
- Aria (warm narrator)
- Nova (bright, upbeat)
- Luna (soft storytelling)
- Sage (calm, instructional)
- Felix (energetic promo)
- Milo (friendly, conversational)
23 supported languages
Support includes major languages like English, Spanish, and Japanese, as well as less common options like Malay and Swahili. If the language is not explicitly set, the system guesses from the script, defaulting to English for Latin text. This broad support ensures your app works globally without complex language packs.
- Arabic, Chinese, Danish, Dutch
- English, Finnish, French, German
- Greek, Hebrew, Hindi, Italian
- Japanese, Korean, Malay, Norwegian
- Polish, Portuguese, Russian, Spanish
- Swahili, Swedish, Turkish
Simple JSON request body
Integration is straightforward: send a standard HTTP POST request with a JSON body containing the text, voice, and format. The request shape mirrors the OpenAI speech endpoint, allowing you to use existing OpenAI SDKs by simply changing the base URL to our API endpoint.
Pricing Model
Unlike traditional TTS services that charge by audio duration, our API bills you strictly for the input text length. This model eliminates the uncertainty of variable speech rates, ensuring your costs remain predictable regardless of whether the voice speaks slowly or rapidly.
Pay per character, not per minute
You are charged based on the exact number of characters in your request, including spaces and punctuation. This direct correlation between input and cost simplifies budgeting for high-volume applications, as you only pay for the text you actually send.
$0.01 per 1,000 characters
The standard rate is $10 per 1 million characters, which breaks down to just $0.01 per 1,000 characters. This flat rate applies equally to both the "tts-1" and "tts-1-hd" models, which use the same engine and price.
Free failed requests
If a synthesis request fails due to server issues or invalid parameters, the charge is returned in full. You never pay for audio that is not successfully delivered, protecting your balance from unexpected API errors or network interruptions.
Integration Ease
For indie developers already using the OpenAI ecosystem, integration is seamless. By simply swapping the base URL to https://api.playv57.com/v1, your existing code works without modification. This compatibility extends to other standard SDKs, allowing you to leverage familiar patterns for authentication and request handling.
OpenAI SDK compatible
Use the official OpenAI SDKs by pointing them to our API base. This eliminates the need for custom wrappers or learning a new library. Your existing TypeScript or Python scripts can generate speech with minimal code changes, treating our endpoint as a drop-in replacement for standard speech endpoints.
- Works with official OpenAI SDKs
- Drop-in replacement for standard speech endpoints
- Supports TypeScript and Python
Standard HTTP POST endpoint
Send a JSON body to POST https://api.playv57.com/v1/audio/speech. The request shape matches the OpenAI speech endpoint, accepting fields like model, input, voice, and response_format. This standardization means you can use any HTTP client or library that supports JSON POST requests.
- POST to /v1/audio/speech
- JSON body with standard fields
- Compatible with any HTTP client
No complex setup required
Start generating audio immediately after obtaining your API key. There are no complex configurations, SDK downloads, or binary dependencies. Just make the request, receive the audio file, and integrate it into your web app. This simplicity reduces development time and keeps your stack lightweight.
- Instant start with API key
- No SDK downloads or binaries
- Lightweight integration

Voice Variety
Choosing the right voice is critical for user experience. Our eight presets cover distinct tonal ranges, from warm narrators to energetic promo styles. Each voice is grounded in licensed recordings, ensuring clarity and consistency across all 23 supported languages.
You can match the tone to your content precisely. Whether you need a calm instructional guide or a deep documentary narration, the API delivers reliable results without requiring custom voice cloning or SSML, which is not supported.
Warm narrator tones
Aria offers a clear, warm narration style ideal for storytelling. Luna provides a soft, engaging voice perfect for long-form content or guided experiences. These voices maintain natural cadence, making them suitable for audiobooks or educational materials.
- Aria: Warm, clear narrator
- Luna: Soft, storytelling voice
Energetic promo styles
Felix delivers an upbeat, energetic tone perfect for promotional content or short ads. Nova adds a bright, lively quality that keeps listeners engaged. These voices are optimized for short bursts of text where energy matters more than depth.
- Felix: Energetic, promo style
- Nova: Bright, upbeat voice
Neutral news delivery
Atlas provides a neutral, professional tone suitable for news updates or factual content. Milo offers a friendly, conversational style for customer support or interactive guides. Orion’s deep, documentary tone works well for serious or informative content.
- Atlas: Neutral, news style
- Milo: Friendly, conversational
- Orion: Deep, documentary voice
Data Privacy
When you send text to our API, the input is used strictly to produce the audio file. We do not use your source text to train our models or improve the engine. This ensures your proprietary content remains yours, avoiding the risk of your phrases appearing in other users' outputs or being ingested into future model versions.
After synthesis completes, the generated audio is returned in the response and is not stored on our servers. Only request metadata—such as timestamp, character count, and status—is retained, and solely for billing purposes. This minimal retention policy means your audio files are ephemeral, reducing long-term data exposure.
Text not used for training
Your input text is processed exclusively for the immediate generation of the audio file. It is not used for training, ensuring your unique content does not influence future model iterations or appear in other users' results.
This approach is critical for indie developers handling sensitive or proprietary content, as it guarantees that your voiceovers do not become part of the underlying engine's memory.
Audio not stored after generation
Once the audio file is delivered to your client via the HTTP response, it is not persisted in our storage systems. Unlike services that archive your generated voices for later access, we treat each request as a transient transaction.
This reduces storage costs and limits the attack surface for data breaches, as there are no historical audio files to leak or subpoena.
Metadata kept for billing
We retain only the essential metadata required to calculate your usage and verify charges: the time of the request, the character count, and the success status. This data is used strictly for accounting and does not include the actual text or audio content.
This minimal logging ensures transparency in your billing while maintaining a lean data footprint, aligning with the privacy expectations of modern web applications.
Performance
Synthesis speed typically matches real-time playback, generating roughly 25 characters of text per second for standard-length inputs. While short phrases require a few seconds to process, longer texts maintain a consistent generation rhythm without significant delays.
Real-time generation speed
For most use cases, the time to generate audio aligns closely with the duration of the resulting speech. This predictable cadence allows developers to estimate processing times accurately based on input length.
Queue handling under load
Requests are processed one after another. If multiple clients submit synthesis jobs simultaneously, additional requests may wait in a queue until previous tasks complete. This single-threaded approach ensures consistency but can introduce latency during peak usage periods.
60 requests per minute limit
Each API key is capped at 60 requests per minute once funded. New accounts start with a lower limit of 10 requests per minute until the first top-up is processed. This rate limit helps maintain stable generation speeds for all users.
Getting Started
Access the /get-api-key/ page to create your account. You can sign in instantly using your Google account or register with an email and password. Each account is tied to this specific storefront.
Once registered, your dashboard displays your active API key, current balance, and usage statistics. You can also generate a new key at any time, which immediately invalidates the previous one.
Sign up with Google or email
Registration takes seconds. Google sign-in provides immediate access, while email registration requires basic credential setup. There are no complex verification steps or waiting periods.
Get your API key instantly
Your unique key appears on the dashboard immediately after creation. Use this key in the Authorization header for all API requests. You can issue a replacement key anytime, ensuring security without downtime.
- Key format: tts-...
- One active key per account
- Instant invalidation of old keys
Start with free trial credit
Google sign-ups receive $0.50 in trial credit, valid for 7 days. This allows you to generate approximately 50,000 characters to test the API. Email-based accounts start with a zero balance and require a top-up.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Creating games alone is hard work but the freedom to follow your creative vision makes every late night worth it.
23 languages
Any voice speaks any of these languages.
