Cloud Text-to-Speech API for Audiobook Production

By Audibook editorial team · Updated 2026-11-18

Audibook provides a reliable cloud text-to-speech API engineered for audiobook production and self-publishing workflows. By leveraging licensed voice recordings from LibriVox, we deliver high-fidelity synthetic speech that publishers can integrate directly into their content pipelines without managing infrastructure.

This page outlines the core capabilities of our TTS endpoint, including voice presets, language support, and technical limits. You will find concrete details on pricing, data privacy, and integration steps, helping you evaluate if our character-based API fits your audio production requirements.

Cloud Text-to-Speech API for Audiobook Production

Frequently asked questions

Is the voice generator truly free?

No, the service is not free. Pricing is $10 per 1 million characters. Failed requests are free, but successful synthesis incurs a charge. There are no monthly plans or subscriptions; you pay only for the characters you generate.

How do characters count in TTS?

Characters include spaces and punctuation. The rate is $10 per 1 million characters, or $0.01 per 1,000 characters. Each request supports up to 4,096 characters. You can split longer texts into parts and join the resulting audio files to handle full-length books.

Can I use the API for commercial audiobooks?

Yes. The API is designed for scalable audio content creation. Input text is used only to produce audio and is not used for training. Generated audio is returned in the response and not stored, allowing you to own the resulting files for commercial distribution.

What happens if the API 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 during production workflows where occasional errors might occur.

How are long books handled?

Split texts into parts of up to 4,096 characters per request. Generate audio for each part, then join the files. Typical speed is roughly real time (25 characters per second). Requests are processed sequentially, so queueing may occur under load.

Do voices sound robotic?

Voices are based on licensed recordings from LibriTTS-R and LibriVox public-domain audiobook readings. They are high-fidelity synthetic voices. Every file carries an inaudible watermark marking it as synthetic speech, but they are designed for natural audiobook narration without obvious robotic artifacts.

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 supported language. Vietnamese, Indonesian, and Thai are not supported.

Is my text data stored?

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. We keep request metadata (time, characters, status) for billing purposes only.

Can I clone a voice?

No. We do not offer custom voice cloning. You cannot upload your own voice or design custom voices. You must choose from the 8 preset voices (Aria, Nova, Luna, Sage, Felix, Milo, Atlas, Orion) available in the API.

Is there an SLA guarantee?

No. We do not provide an SLA or uptime guarantee. There are no enterprise contracts or certifications like SOC 2 or HIPAA. Support is by email only. Requests are processed sequentially, so latency may vary under load.

Do you support SSML?

No. The API does not support SSML, phoneme dictionaries, or custom pronunciation lexicons. You control speech via the speed parameter (0.25–4.0) and voice selection. Emotion tags and SSML markup are not available in the request body.

Can I stream audio in real-time?

No. There is no streaming, WebSocket, or real-time partial audio. You send a POST request and receive the complete audio file in the response. The endpoint returns the full synthesized file once processing is complete.

How do I pay with USDT?

Top-up 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, or invoices. Balance does not expire and has no minimum spend.

Is there a free trial?

Yes. Signing in with Google provides $0.50 of credit (about 50,000 characters) valid for 7 days. Accounts created with email and password start without trial credit. You can top up with USDT to continue using the service after the trial.

First request

curl

curl https://api.playtk66.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.mp3

API Overview

Sound wave illustration for: API Overview

Audibook provides a cloud-based text-to-speech API engineered for audiobook production. The endpoint accepts JSON payloads and returns audio files, integrating directly into self-publishing workflows without infrastructure overhead. Requests are processed sequentially, ensuring consistent delivery of synthetic speech with transparent character-based billing.

The service is compatible with the OpenAI SDK structure. Developers can route requests to https://api.playtk66.com/v1 using standard libraries, enabling immediate adoption for existing audio generation pipelines. This compatibility simplifies integration for publishers scaling content creation.

Cloud-based TTS endpoint

The primary endpoint is POST https://api.playtk66.com/v1/audio/speech. It accepts text input and returns generated audio files in specified formats. The API handles voice synthesis on cloud servers, eliminating the need for local processing power or specialized hardware.

Requests are processed one after another. Under load, requests may wait in a queue. Typical synthesis speed is roughly real time for long texts, with short phrases taking a few seconds to generate.

  • Endpoint: POST https://api.playtk66.com/v1/audio/speech
  • Input: JSON body with text and voice parameters
  • Output: Audio file in selected format

OpenAI-compatible SDK

The API mirrors the request shape of the OpenAI speech endpoint. This allows developers to use official OpenAI SDKs by simply changing the base_url to https://api.playtk66.com/v1 and supplying the site's API key. No custom client libraries are required.

This compatibility ensures that standard authentication and payload structures work out of the box. Developers familiar with OpenAI's integration pattern can migrate their code with minimal adjustments.

JSON request structure

Requests are sent as JSON objects containing the text, voice selection, and format preferences. The model field accepts "tts-1" or "tts-1-hd", both using the same engine and pricing tier. The input field holds the text, limited to 4096 characters per request.

Optional fields include language for explicit ISO code specification, speed for rate adjustment (0.25–4.0), and response_format for output type. Authentication is handled via the Authorization: Bearer header with the user's API key.

Supported Voices

Sound wave illustration for: Supported Voices

The API provides eight distinct voice presets engineered for narrative clarity and consistent pacing, suitable for extended audiobook sessions. Each voice offers a specific tonal quality, from warm narration to neutral reporting, allowing publishers to match the voice to the genre without re-recording.

Female presets: Aria, Nova, Luna, Sage

Aria delivers a warm, clear narrator tone ideal for fiction. Nova offers a bright, upbeat delivery for lighter reads. Luna provides a soft, storytelling atmosphere, while Sage ensures calm, instructional clarity for non-fiction.

Male presets: Felix, Milo, Atlas, Orion

Felix brings energetic delivery suitable for promotional content or fast-paced narratives. Milo provides a friendly, conversational tone. Atlas maintains a neutral, news-like delivery, and Orion offers a deep, documentary-style voice for serious or historical texts.

Licensed source material

All voices are based on licensed recordings from LibriTTS-R (CC BY 4.0), derived from LibriVox public-domain audiobook readings. This foundation ensures high-fidelity synthetic speech with consistent audio quality. Every generated file includes an inaudible watermark to mark it as synthetic speech.

Supported Voices

Language Support

Sound wave illustration for: Language Support

Our TTS engine supports 23 languages, covering major global markets including Arabic, Chinese, English, French, German, Hindi, Japanese, Korean, Russian, Spanish, and Turkish. Any voice can speak any supported language, providing flexibility for diverse audiobook catalogs.

The API automatically detects language from the input script, defaulting to English for Latin text. For non-English Latin-script text, explicitly setting the language parameter ensures accurate pronunciation and avoids misinterpretation.

23 supported languages

We support 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 currently supported.

Automatic language detection

If the language field is omitted, the system infers the language from the character script. Latin script defaults to English; other scripts trigger detection based on common patterns.

Explicit language setting

For non-English Latin-script text (e.g., Spanish or Portuguese), set the language ISO code explicitly. This prevents default English pronunciation and ensures accurate voice rendering.

  • Set language for non-English Latin text
  • Omit for automatic script detection
  • Unsupported languages return an error

Audio Formats

Sound wave illustration for: Audio Formats

The API returns audio in standard formats via the response_format parameter. Choose the output type that best fits your encoding pipeline, whether you need universal compatibility or lossless quality for archival masters.

Each format is generated from the same high-fidelity engine, ensuring consistent voice characteristics regardless of the container format selected for your audiobook distribution.

MP3 output

MP3 is the default and most widely supported format for audiobook platforms. It offers a balance between file size and audio quality, making it ideal for streaming and general distribution.

Use mp3 for efficient bandwidth usage while maintaining clear speech reproduction suitable for most consumer listening devices.

WAV output

WAV provides uncompressed PCM audio. This format is suitable for editing workflows or platforms requiring lossless input.

Select wav when you need to apply further post-processing or when your distribution channel mandates uncompressed audio data.

Opus, FLAC, PCM support

For specialized use cases, the API supports Opus, FLAC for lossless compression, and raw PCM for direct digital audio processing.

Choose these formats based on your specific infrastructure requirements, such as minimizing storage costs with Opus or preserving bit-perfect audio with FLAC.

  • Opus: Efficient streaming
  • FLAC: Lossless compression
  • PCM: Raw uncompressed data

Pricing Model

Sound wave illustration for: Pricing Model

Our pricing is strictly character-based at $0.01 per 1,000 characters, including spaces and punctuation. This transparent model eliminates hidden fees, allowing publishers to calculate production costs precisely before synthesis begins.

$0.01 per 1,000 characters

Each character in your input text counts toward the total. For a standard 50,000-character audiobook chapter, the cost is exactly $0.50. This predictable rate applies to all models and voices, simplifying budget forecasting for self-publishing workflows.

Failed requests are free

If synthesis fails, the charge is returned in full. You only pay for successfully generated audio files, ensuring you are never billed for errors in processing or temporary service interruptions.

No minimum spend

There are no monthly plans, seats, or minimum spending requirements. Balance does not expire, and you can top up via USDT on the TRON network whenever needed. Start with the free trial credit or pay only for what you use.

Pricing Model

Technical Limits

Sound wave illustration for: Technical Limits

Efficient batch processing requires splitting long manuscripts into manageable segments. The API enforces a strict 4096-character limit per request, ensuring consistent synthesis quality and predictable resource usage for each audio chunk.

4096 character limit per request

Each API call accepts a maximum of 4096 characters, including spaces and punctuation. Text exceeding this length must be segmented by the client application before submission.

  • Limit: 4096 characters (including spaces/punctuation) per request
  • Longer texts require client-side splitting and audio joining

60 requests per minute

Standard rate limits allow 60 requests per minute per API key. New accounts without a top-up are restricted to 10 requests per minute, suitable for initial testing and small-scale publishing batches.

  • Standard limit: 60 requests/minute
  • Pre-top-up limit: 10 requests/minute

Queue processing under load

Requests are processed sequentially. During high demand, synthesis may wait in a queue before beginning. Generation speed averages roughly 25 characters per second, meaning longer segments take proportionally more time to complete.

Data Privacy

Sound wave illustration for: Data Privacy

Input text is consumed solely for the synthesis process and is never used to train or improve the underlying voice models. This ensures that your manuscript content remains distinct from the model’s training data, preserving the integrity of your specific publishing assets.

Text not used for training

The raw text submitted to the API is processed exclusively to generate the requested audio file. It is not ingested into the learning dataset used to refine voice characteristics or language patterns.

Audio not stored

Generated audio files are returned directly in the API response and are not persistently stored on our servers after delivery. This minimizes the risk of unauthorized access to your finished audiobook chapters or demos.

Metadata kept for billing

We retain request metadata including timestamp, character count, and status code. This data is used strictly to calculate charges based on your usage and to verify the success of each synthesis job.

Usage logs are maintained to ensure accurate billing and to support troubleshooting if a request fails or returns an unexpected status code.

  • Timestamp of request
  • Character count processed
  • Request status (success/fail)
  • Voice and model used

Getting Started

Sound wave illustration for: Getting Started

Create an account on the /get-api-key/ page using Google or email. This grants immediate access to your unique API key and a $0.50 free trial valid for seven days.

Your key authenticates all requests. You can revoke and issue a new key at any time; the old one stops instantly. No subscriptions or monthly seats are required.

Sign up with Google

Visit the sign-up page and select Google authentication. This method automatically applies a $0.50 trial credit (approx. 50,000 characters) valid for 7 days.

  • Instant account creation
  • Trial credit applied automatically
  • 7-day validity period

Receive API key

Upon registration, your dashboard displays your active API key. Use this key in the Authorization header for all API calls.

  • Single active key per account
  • Revoke and regenerate instantly

First curl request

Test your key with a simple synthesis request. Replace YOUR_API_KEY with your actual key.

  • curl -X POST https://api.playtk66.com/v1/audio/speech \
  • -H "Authorization: Bearer YOUR_API_KEY" \
  • -H "Content-Type: application/json" \
  • -d '{"model": "tts-1", "input": "Hello world", "voice": "nova", "response_format": "mp3"}'
Getting Started

Use Cases

Sound wave illustration for: Use Cases

Transform text manuscripts into professional-grade audio without hiring voice talent or managing recording studios. Our API handles the entire synthesis pipeline, delivering consistent narration quality across thousands of titles with predictable character-based pricing.

Audiobook production

Scale your backlist from print to audio instantly. Use distinct voices like Atlas for news or Luna for storytelling to maintain brand consistency across series without re-recording.

  • Consistent voice across multiple books
  • No studio booking overhead
  • Instant generation for new releases

Self-publishing audio

Publishers can generate audiobooks directly from source files. The transparent pricing model ensures costs remain fixed regardless of production volume, enabling profitable marginal units.

  • Direct text-to-audio workflow
  • Transparent character-based billing
  • No minimum volume requirements

Long-text splitting

Automatically split manuscripts exceeding 4,096 characters into sequential requests. Split texts into parts of up to 4,096 characters and join the audio files.

  • Automatic chunking up to 4096 chars
  • Seamless audio concatenation
  • Handles long-form content efficiently

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. Empower authors to share their stories and build lasting careers through direct reader engagement.

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. Python documentation
  3. MDN: Web Speech API
  4. Creative Commons Attribution 4.0
  5. LibriTTS-R corpus (OpenSLR 141)
  6. openai-python on GitHub
  7. TRON developer hub: TRC-20
  8. W3C WAI: making audio and video accessible
  9. Node.js documentation
  10. OpenAI API reference: create speech
  11. IETF RFC 7845: Ogg Opus
  12. LibriTTS-R paper (arXiv 2305.18802)

More on this site