Cloud Text-to-Speech API for Indie Developers

By Lexiconet editorial team · Updated 2026-11-18

This guide explains how indie developers and product teams can turn written content into natural-sounding audio without building complex infrastructure. We offer a simple, cost-effective API that integrates with your existing workflows using standard OpenAI SDKs, letting you focus on your product rather than audio engineering.

You will find clear details on our 8 licensed voices, 23 supported languages, and straightforward pay-per-character pricing. Whether you are building a podcast app, an e-learning tool, or an accessibility feature, this page provides the technical specifics and integration steps you need to get started quickly and reliably.

Cloud Text-to-Speech API for Indie Developers

Frequently asked questions

What is the pricing model?

We charge $10 per 1 million characters, which equals $0.01 per 1,000 characters. Spaces and punctuation count toward this total. Failed requests are free, with the charge returned in full. There are no monthly plans, seats, or minimum spend requirements, making it simple and predictable for indie developers.

Which languages are supported?

We support 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 you do not set the language, it is guessed from the script.

Is the API compatible with OpenAI SDK?

Yes. Our endpoint matches the OpenAI speech endpoint request shape. You can use the official OpenAI SDKs by changing the base_url to https://api.playf65.com/v1 and using your site-specific key. This allows you to integrate our TTS service into your existing workflows with minimal code changes.

How do I top up my balance?

Top up using USDT on the TRON network (TRC-20) only. The minimum top-up amount is $5. Credits are automatically added to your account within minutes after the transfer is confirmed. We do not accept credit cards, PayPal, or subscriptions, keeping the process straightforward and decentralized.

Are there any hidden fees?

No. We charge strictly by character usage at $10 per 1 million characters. Failed requests are free, and the charge is returned in full. There are no monthly plans, no seat fees, and no hidden costs. Your balance does not expire, so you only pay for what you use.

Can I use my own voice?

No. We do not offer custom voice cloning or allow you to upload your own voice. You can choose from eight preset voices based on licensed recordings from LibriTTS-R and LibriVox. Each generated file includes an inaudible watermark to mark it as synthetic speech, ensuring transparency.

Is SSML supported?

No. Our API does not support SSML, phoneme dictionaries, or custom pronunciation lexicons. You send plain text up to 4096 characters per request. If you need precise control over pronunciation or pacing beyond the speed parameter, you may need to split your text and adjust speeds manually.

How long does generation take?

Generation is roughly real-time, producing about 25 characters of text per second of audio. Short phrases may take a few seconds. Requests are processed sequentially, so under heavy load, a request might wait in a queue. For long texts, split them into parts and join the audio files.

Is my data used for training?

No. Your input text is used only to produce the audio and is not used for training models. Generated audio is returned in the response and not stored. We keep request metadata, such as time, character count, and status, solely for billing purposes and account management.

What happens if a request fails?

If synthesis fails, the charge is returned in full. You are not billed for failed requests. This ensures you only pay for successful audio generation. You can retry the request.

Why Choose Our API

Sound wave illustration for: Why Choose Our API

We built this API for indie developers who need reliable speech synthesis without complex infrastructure or enterprise overhead. Our solution strips away the noise, offering a straightforward integration that lets your product team focus on building features rather than managing audio pipelines. You get high-quality output with minimal code changes.

Simple pay-per-character pricing

You only pay for what you use. At $0.01 per 1,000 characters, costs remain predictable and low. Failed requests are free, so you never pay for errors. Top-ups via USDT are instant, and your balance never expires.

  • $0.01 per 1,000 characters
  • Free failed requests
  • No monthly plans or minimum spend
  • Instant USDT top-ups

8 licensed voices, 23 languages

Choose from eight distinct presets based on licensed recordings, supporting 23 languages. Each voice has a unique tone, from warm narration to energetic promos. All voices can speak any supported language, ensuring consistent quality across your global user base.

  • 8 distinct voice presets
  • Support for 23 languages
  • Licensed recordings from LibriVox
  • No voice cloning required

OpenAI SDK compatible

Use the official OpenAI SDK with a simple base_url change. Our API follows the same request shape, allowing you to drop it into existing projects with minimal friction. Standard JSON endpoints and clear documentation make integration fast and reliable.

  • Drop-in OpenAI SDK compatibility
  • Standard JSON endpoints
  • Easy curl examples
  • No custom SDK needed

Pricing and Payment

Sound wave illustration for: Pricing and Payment

Our pricing is straightforward: you pay $0.01 per 1,000 characters, including spaces and punctuation. Failed requests are free, so you only pay for successful synthesis. This model scales naturally with your usage, avoiding complex tiered structures or minimum spend requirements.

$0.01 per 1,000 characters

Costs are calculated per character, making it easy to predict expenses for projects of any size. With a typical speed of roughly real-time generation, long texts process efficiently. You retain full control over your budget without unexpected overages.

USDT (TRC-20) top-ups only

We support USDT on the TRON network (TRC-20) for top-ups. Minimum deposit is $5, and credits are automatically added within minutes. We do not process cards, PayPal, or invoices, keeping the payment process simple and global.

  • No monthly plans or subscriptions
  • Balance never expires
  • Automatic credit within minutes

Free trial credit included

Sign in with Google to receive $0.50 in trial credit (about 50,000 characters), valid for 7 days. Email-based accounts start without this credit. Use it to test voices and integration before committing to a top-up.

Voices and Languages

Sound wave illustration for: Voices and Languages

Our API provides eight distinct voice presets, ranging from warm narrators to energetic promo voices. Each voice is based on licensed recordings from LibriVox, ensuring a natural, human-like quality without the need for complex engineering on your part. You can choose from female and male options to match your content's tone perfectly.

These voices support 23 languages, including English, Spanish, German, and Japanese. Any voice can speak any of these languages, giving you flexibility in your multilingual projects. This wide support allows indie developers to reach global audiences without managing multiple voice models.

8 distinct voice presets

We offer eight licensed voice presets based on LibriTTS-R recordings. Female voices include Aria (warm narrator), Nova (bright and upbeat), Luna (soft storytelling), and Sage (calm instructional). Male voices include Felix (energetic promo), Milo (friendly conversational), Atlas (neutral news), and Orion (deep documentary).

Each voice has a distinct tone, allowing you to select the one that fits your content. Whether you need a friendly conversational tone or a deep documentary style, we have a voice that matches your needs.

  • Aria: Warm, clear narrator
  • Nova: Bright, upbeat
  • Luna: Soft, storytelling
  • Sage: Calm, instructional
  • Felix: Energetic, promo
  • Milo: Friendly, conversational

Support for 23 languages

Our API supports 23 languages, including 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, giving you flexibility in your multilingual projects.

If you do not set the language, the system guesses it from the script. Latin text defaults to English, so for non-English Latin-script text, you should set the language manually to ensure accurate pronunciation.

  • Arabic
  • Chinese
  • Danish
  • Dutch
  • English
  • German

No voice cloning or SSML

We focus on simplicity and reliability. Our API does not support voice cloning, SSML, phoneme dictionaries, or custom pronunciation lexicons. You cannot upload your own voice or design custom emotions. This keeps the integration straightforward and the pricing predictable.

Every generated file carries an inaudible watermark that marks it as synthetic speech. This ensures transparency about the use of AI-generated audio in your content.

Integration

Sound wave illustration for: Integration

Integrating our TTS API is straightforward for indie developers who already use the OpenAI ecosystem. By simply pointing your SDK’s base_url to https://api.playf65.com/v1 and using your unique API key, you can utilize the official OpenAI libraries without code changes. This compatibility ensures a smooth transition for existing workflows while keeping your infrastructure costs low.

OpenAI SDK compatible

We mirror the standard request shape, allowing you to send text and receive audio files using familiar libraries. This means you can leverage the robust OpenAI SDKs for Python, Node.js, and other languages, changing only the endpoint configuration. You get a reliable, developer-friendly experience without learning a new client library.

  • Use official OpenAI SDKs by updating the base URL
  • Same JSON request body structure as OpenAI’s speech endpoint
  • Works with Python, JavaScript, and other major languages

Standard JSON endpoints

Our API uses standard HTTP methods and JSON payloads, making it easy to integrate into any backend service. The primary endpoint, POST https://api.playf65.com/v1/audio/speech, accepts your text, voice selection, and audio format preferences. You can also retrieve available voices and models via GET endpoints, ensuring you have all the metadata needed for dynamic generation.

  • POST for speech generation
  • GET /v1/voices for voice and language metadata
  • GET /v1/models for model information

Easy curl examples

For quick testing, you can use standard curl commands to verify your integration. This is particularly useful for debugging or initial setup before diving into your application code. Simply replace the placeholder API key with your own, and you will receive an audio file in your chosen format within seconds.

  • Quick verification using command-line tools
  • Minimal setup required for initial tests
  • Immediate feedback on request structure

Comparison with Others

Sound wave illustration for: Comparison with Others

Choosing the right TTS provider often means balancing cost, complexity, and ease of integration. For indie developers, overhead matters. Our API strips away enterprise bloat while keeping the developer experience smooth and predictable.

Lower cost than OpenAI

OpenAI charges $15 per million characters for tts-1 and $30 for tts-1-hd. We charge $10 per million characters for both models. The pricing is flat, with no tiered jumps or hidden fees for higher volume. You pay exactly for what you generate, making budgeting straightforward for small teams.

  • OpenAI: $15–$30 per 1M characters
  • Us: $10 per 1M characters
  • Same engine quality, lower base rate

Simpler than Google Cloud

Google Cloud Text-to-Speech requires managing complex projects, IAM roles, and multiple API keys across different services. Our API is a single endpoint. You send a POST request with a JSON body, and you receive an audio file. No service configuration, no complex authentication chains. Just key, text, voice, and audio.

  • Single API key per account
  • No project configuration needed
  • Direct JSON input, direct audio output

Less complex than Amazon Polly

Amazon Polly offers numerous neural voices, SSML tags, and configuration options that can slow down integration. We provide eight clear presets that cover most use cases. You choose a voice, set a speed, and go. If you need SSML, you can add it, but our API works perfectly without it, keeping your requests lightweight.

  • 8 curated voices vs. dozens of options
  • No mandatory SSML learning curve
  • Cleaner request payload structure

Technical Details

Sound wave illustration for: Technical Details

Each API key is limited to 60 requests per minute, though this starts at 10 requests per minute before your first top-up. Requests are processed sequentially, so under heavy load a single request may wait in a queue. Generation speed is roughly real-time, producing about 25 characters of text per second of audio.

Rate limits and queueing

Longer texts must be split into chunks of up to 4096 characters and joined after generation. Short phrases typically take a few seconds to process. If synthesis fails, the request is not charged, ensuring you only pay for successful outputs.

  • 60 requests/minute per key (10 before first top-up)
  • Sequential processing with queueing under load
  • Failed requests are free

Audio formats supported

You can choose from mp3, wav, opus, flac, or pcm via the response_format field. There are no extra fees for format selection, allowing you to optimize for file size or playback compatibility without additional cost.

  • mp3, wav, opus, flac, pcm
  • No extra fees for format selection

Data privacy and usage

Your input text is used solely to produce the audio and is never used for training models. Generated audio is returned in the response and not stored on our servers. We retain request metadata, such as time and character count, strictly for billing purposes.

  • Input text not used for training
  • Generated audio not stored
  • Metadata kept for billing

FAQ

Sound wave illustration for: FAQ

Answering your most frequent questions about reliability, cost, and capabilities to help you integrate quickly.

Is it cheaper than OpenAI?

Our pricing is $0.01 per 1,000 characters, which is lower than OpenAI's standard rates. This flat rate makes budgeting predictable for indie developers without hidden tier costs.

Which languages are supported?

We support 23 languages including English, Spanish, German, Japanese, and Arabic. If the language code is omitted, the system guesses from the script, defaulting to English for Latin text.

  • Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, and Bengali are not supported.

How do I pay?

Payments are processed via USDT (TRC-20) on the TRON network. Minimum top-up is $5, and funds credit automatically within minutes. We do not accept credit cards or PayPal.

Can I clone voices?

No. We offer eight preset voices based on licensed recordings. We do not provide custom voice cloning, uploading your own voice, or emotion tags.

Is there a free tier?

Google sign-ups receive $0.50 trial credit (approx. 50,000 characters) valid for 7 days. Email sign-ups start with zero balance. Failed requests are always free.

FAQ

Supported Audio Formats

Sound wave illustration for: Supported Audio Formats

Generate audio in five distinct formats: MP3, WAV, Opus, FLAC, and PCM. Each file is delivered directly in the API response, ensuring you receive the exact binary data needed for your application without intermediate processing steps.

MP3, WAV, Opus, FLAC, PCM

Select the format that best matches your delivery infrastructure. MP3 offers efficient compression for web streaming, while WAV provides uncompressed quality for local playback. Opus balances file size and audio fidelity, whereas FLAC ensures lossless preservation. PCM delivers raw audio data for immediate processing.

  • MP3: Optimized for web delivery and streaming
  • WAV: High-quality uncompressed audio for local playback
  • Opus: Efficient compression with excellent fidelity
  • FLAC: Lossless compression for archival quality
  • PCM: Raw audio data for immediate processing

Choose the format for your needs

Consider your endpoint requirements when selecting a format. Web applications often benefit from MP3 or Opus to reduce bandwidth usage, while desktop or mobile apps may prefer WAV for compatibility. Use PCM if you need raw samples for further audio manipulation.

No extra fees for format selection

The price remains consistent regardless of the format you choose. Whether you generate 100 KB MP3 files or 1 MB WAV files, you are charged based on the input character count, not the output file size. This simplifies cost estimation and budgeting.

Voice Presets Overview

Sound wave illustration for: Voice Presets Overview

Choose from eight voice presets tailored for indie development needs. Our collection includes four female and four male voices, each recorded from licensed sources and optimized for clarity in API-driven applications.

Female voices: Aria, Nova, Luna, Sage

Aria offers a warm, clear narrator style ideal for audiobooks. Nova provides a bright, upbeat tone perfect for short promos. Luna delivers a soft, storytelling voice for immersive content. Sage ensures calm, instructional clarity for tutorials.

  • Aria: Warm, clear narrator
  • Nova: Bright, upbeat
  • Luna: Soft, storytelling
  • Sage: Calm, instructional

Male voices: Felix, Milo, Atlas, Orion

Felix brings an energetic, promo-ready energy. Milo sounds friendly and conversational for customer-facing apps. Atlas delivers a neutral, news-style tone. Orion offers a deep, documentary voice for serious content.

  • Felix: Energetic, promo
  • Milo: Friendly, conversational
  • Atlas: Neutral, news
  • Orion: Deep, documentary

Each voice has a distinct tone and style

Every voice is based on licensed recordings from LibriTTS-R and LibriVox public-domain audiobooks. This ensures consistent quality and distinct character across all generated audio files without requiring custom voice design.

Language Support Details

Sound wave illustration for: Language Support Details

Our API supports 23 languages, enabling you to reach a global audience without managing multiple voice models. You can generate audio in 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 preset can speak any of these languages, giving you flexibility in tone and style across regions.

This broad support ensures your indie app or product can serve diverse user bases effectively. Whether you are localizing a narrative game or creating educational content, the consistent API behavior across languages simplifies integration. You do not need to switch endpoints or learn new parameters for each language.

23 languages supported

The API handles 23 distinct languages, covering major global markets and many regional needs. This list includes English, Spanish, French, German, Japanese, Korean, Chinese, and others like Swahili and Turkish. Each voice preset is capable of speaking any of these languages.

You can mix languages within your application by selecting the appropriate voice and setting the correct language code. This allows for multilingual content without requiring separate voice profiles for each language.

  • 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 you do not specify a language, the API guesses it from the text script. For example, Latin-script text defaults to English. This works well for single-language content but may cause errors if the text contains mixed scripts or non-English Latin characters.

Relying on auto-detection is convenient for quick prototypes. However, for production apps with non-English Latin text, explicit language setting is recommended to ensure accurate pronunciation and tone.

Manual language code setting available

You can set the language field in the request body to an ISO code (e.g., "es" for Spanish). This ensures the API uses the correct phonetic rules for that language, even if the text script is Latin. This is crucial for languages like Turkish that use Latin script but require specific handling.

Using explicit language codes prevents mispronunciation and ensures consistent output quality. It is the recommended approach for any content that is not English.

Account and Setup

Sound wave illustration for: Account and Setup

Creating an account on playf65.com takes seconds. You can sign up using your existing Google account for quick access, or choose a standard email and password setup. Each account is tied to the specific site you visit, ensuring a clean, dedicated workspace for your project.

Sign up with Google or email

Choose the method that fits your workflow. Google sign-in is instant, while email registration gives you full control over your credentials.

Get your API key instantly

Once registered, your unique API key is generated immediately. This key authenticates your requests to the endpoint and can be rotated at any time, revoking the old key instantly if needed.

  • Keys are unique per account
  • Instant rotation available

View balance and usage in dashboard

The dashboard provides real-time visibility into your account status. You can see your current balance, any trial credit remaining, and detailed usage metrics to keep your project costs predictable.

  • Real-time balance tracking
  • Usage metrics overview

Data Privacy and Usage

Sound wave illustration for: Data Privacy and Usage

We prioritize your data’s integrity by ensuring your input text is used exclusively for generating the requested audio file. Unlike some competitors, we do not use your input text to train or improve our underlying models, keeping your content strictly yours.

Once the audio is generated and delivered in the response, it is not stored on our servers. This minimizes long-term data retention risks, ensuring your specific audio outputs remain transient unless you choose to save them locally.

Input text not used for training

Your input text is processed solely to produce the audio file. We do not retain or utilize your text samples for model training purposes, ensuring your creative content remains private and distinct from our training datasets.

Generated audio not stored

The synthesized audio is returned directly in the API response and is not persisted on our servers after delivery. This approach ensures that your generated audio files exist only where you choose to save them.

Metadata kept for billing purposes

We retain request metadata, including timestamp, character count, and status, solely for accurate billing and usage tracking. This minimal data allows us to manage your account balance and provide transparent usage reports without storing the actual content.

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. Use PCM if you need raw samples for further audio manipulation.

Avatar of the voice aria
ariafemale
Avatar of the voice nova
novafemale
Avatar of the voice luna
lunafemale
Avatar of the voice sage
sagefemale
Avatar of the voice felix
felixmale
Avatar of the voice milo
milomale
Avatar of the voice atlas
atlasmale
Avatar of the voice orion
orionmale

23 languages

Any voice speaks any of these languages.

Sources and further reading

  1. Wikipedia
  2. openai-python on GitHub
  3. LibriTTS-R corpus (OpenSLR 141)
  4. Creative Commons Attribution 4.0
  5. Python documentation
  6. FFmpeg documentation
  7. openai-node on GitHub
  8. LibriTTS-R paper (arXiv 2305.18802)
  9. curl documentation
  10. OpenAI API reference: create speech
  11. ISO 639 language codes
  12. W3C WAI: making audio and video accessible

More on this site