Frequently asked questions
What is AI voice cloning online?
AI voice cloning involves creating a digital replica of a specific person’s voice. Our service does not offer custom voice cloning for customers, nor do we allow uploading your own voice. We provide eight fixed presets based on licensed recordings. Custom cloning is not currently available, so you must select from the existing voice library for your audiobook production.
How does synthetic voice synthesis differ from traditional audiobook recording?
Synthetic synthesis generates audio instantly via API requests, eliminating studio time and human narrator fees. Traditional recording requires scheduling, equipment, and editing. Our API delivers high-fidelity audio files directly, enabling scalable production without infrastructure overhead. While traditional methods offer human nuance, our synthetic voices provide consistent quality and predictable billing for self-publishing professionals.
Which voices are best for fiction audiobooks?
For fiction, Aria offers a warm, clear narrator style suitable for general storytelling. Luna provides a soft, storytelling tone ideal for emotional narratives. Orion’s deep, documentary style works well for serious fiction. Nova’s bright, upbeat voice suits lighter fiction. Select based on the specific mood of your novel, as all voices support multiple languages.
Which voices are best for non-fiction or instructional content?
Sage is designed for calm, instructional content, making it ideal for educational or business non-fiction. Atlas provides a neutral, news-style delivery suitable for factual reporting. Felix offers an energetic, promo tone for marketing materials. Milo delivers a friendly, conversational style for accessible non-fiction. These presets ensure clear, professional delivery for informational audiobooks.
Can I use the generated audio for commercial audiobooks?
Yes, the generated audio is suitable for commercial use. The voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain readings. Each file carries an inaudible watermark marking it as synthetic speech. You can scale audio production without infrastructure overhead, making it viable for self-publishing and professional audiobook distribution.
Do the voices sound natural enough for professional publishing?
The voices are based on licensed recordings from LibriTTS-R and LibriVox, ensuring high fidelity. Each generated file carries an inaudible watermark marking it as synthetic speech. While they are synthetic, they are designed for professional use in audiobooks and self-publishing. The quality is consistent, and the clear licensing supports commercial distribution without additional rights management.
How many characters can I process in a single request?
Each API request accepts up to 4096 characters of input text. Spaces and punctuation count toward this limit. For longer texts, you must split the content into parts of up to 4096 characters and join the resulting audio files. This ensures efficient processing and maintains audio quality across the entire audiobook.
What audio formats are supported for delivery?
The API supports five response formats: mp3, wav, opus, flac, and pcm. You can specify the desired format in the request body. MP3 and WAV are common for general distribution, while FLAC offers lossless quality. OPUS and PCM are also available for specific technical requirements or processing pipelines.
Is there a minimum spend requirement?
There is no minimum spend requirement for using the API. You pay only for the characters you consume. The minimum top-up amount is $5 when adding funds via USDT on the TRON network. Balances do not expire, and there are no monthly plans or subscription fees, allowing flexible usage for any volume.
How is the API billed?
Billing is based on character count at $10 per 1 million characters, or $0.01 per 1,000 characters. Failed requests are free. You can top up your balance using USDT on the TRON network, with a minimum of $5. Credits are credited automatically within minutes. There are no invoices, VAT, or monthly plans.
Are failed requests charged?
No, failed requests are free. If synthesis fails, the charge is returned in full. You are only billed for successful audio generation. This ensures you pay strictly for usable output. The system processes requests sequentially, so under load, a request may wait in a queue before being processed or failing.
Can I change my API key if it is compromised?
Yes, you can issue a new API key at any time from your account dashboard. When a new key is issued, the old one stops working immediately. Each account has one active key at a time. This allows you to secure your account quickly if the key is exposed or compromised.
What happens if the service is under high load?
Requests are processed one after another. Under high load, a request may wait in a queue before being synthesized. Typical speed is roughly real time for long texts (about 25 characters per second), but short phrases may take a few seconds. This sequential processing ensures stability but may introduce latency during peak usage periods.
Is there a free trial available for new users?
A free trial of $0.50 credit (about 50,000 characters) is available for 7 days when you sign in with Google. Accounts created with email and password start without trial credit. This allows you to test the API’s quality and integration before committing to a top-up via USDT on the TRON network.
First request
curl
curl https://api.regal33.info/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.mp3Synthetic Voice Technology for Content Creators
BookRead API delivers high-fidelity synthetic voices designed specifically for audiobook producers and self-publishing professionals. By leveraging licensed recordings from LibriTTS-R and LibriVox, we provide clear, natural-sounding audio files that meet professional production standards without the overhead of studio infrastructure.
High-fidelity audio generation
Send text via API and receive audio files in mp3, wav, opus, flac, or pcm formats. Each file includes an inaudible watermark identifying it as synthetic speech, ensuring transparency for distributors and listeners while maintaining broadcast-quality clarity.
Multi-language support
Generate audio in 23 languages, including English, Spanish, German, French, Japanese, and Chinese. Any voice can speak any supported language, allowing you to expand your catalog geographically without hiring separate voice talent for each market.
Consistent tone and quality
Maintain uniform narration across your entire library. With eight distinct presets ranging from warm narrators to neutral news readers, you can select a single voice for a series or match specific tones to different genres, ensuring a cohesive listening experience.
Text-to-Speech API Explained
BookRead API converts written text into natural-sounding audio through cloud-based synthesis. You send a request with your manuscript text and preferred voice, and receive a standardized audio file. This process eliminates the need for physical studios, allowing you to generate audiobooks directly from your digital files.
The technology uses licensed recordings to ensure clarity and consistency. Unlike human narration, which requires scheduling and studio time, the API processes requests sequentially. This allows for immediate generation of audio files, streamlining the production workflow for self-publishing professionals who need reliable output without infrastructure overhead.
Understanding synthetic voice synthesis
Synthetic voice synthesis generates audio from text using pre-recorded voice datasets. Our engine processes your input text and returns an audio file in formats like MP3 or WAV. Each file contains an inaudible watermark to identify it as synthetic speech, ensuring clear rights management for your audiobook productions.
Difference from traditional recording
Traditional audiobook production requires hiring narrators, booking studio time, and managing audio engineering. Our API replaces this linear process with a direct digital conversion. You control the voice selection, speed, and language directly through the API parameters, without coordinating human schedules or dealing with studio logistics.
- No studio booking or engineer fees
- Instant audio generation
- Direct control over voice parameters
Advantages for self-publishing
Self-publishing authors benefit from predictable billing and scalable production. You pay only for the characters you use, with failed requests refunded automatically. This model allows you to generate entire catalogs without upfront costs for narration teams, making it easier to test different voices for your titles.
- Pay-per-character billing
- Automatic refunds for failures
- Scalable production volume
Voice Selection for Audiobooks
Selecting the right voice ensures your audiobooks match the intended atmosphere and listener expectations. BookRead API offers eight distinct presets, each derived from licensed recordings, allowing you to maintain consistent quality across your catalog.
Each voice is optimized for specific narrative styles, from warm narration to energetic promotional reads. This structured selection simplifies production workflows for self-publishing professionals.
Female presets: Aria, Nova, Luna, Sage
Aria provides a warm, clear narrator tone suitable for general fiction. Nova offers a bright, upbeat delivery ideal for contemporary romances or light reads. Luna delivers a soft, storytelling voice, perfect for emotional narratives. Sage provides a calm, instructional tone, well-suited for non-fiction or educational content.
- Aria: Warm, clear narrator
- Nova: Bright, upbeat
- Luna: Soft, storytelling
- Sage: Calm, instructional
Male presets: Felix, Milo, Atlas, Orion
Felix brings an energetic, promo-style delivery, excellent for thrillers or fast-paced plots. Milo offers a friendly, conversational voice, ideal for personal development or memoirs. Atlas provides a neutral, news-style tone for objective storytelling. Orion delivers a deep, documentary voice, suitable for serious historical or scientific works.
- Felix: Energetic, promo
- Milo: Friendly, conversational
- Atlas: Neutral, news
- Orion: Deep, documentary
Matching voice to genre and audience
Align voice characteristics with your target demographic. Use Luna for intimate fiction or Atlas for objective non-fiction. Consistent voice selection across a series enhances listener retention and brand recognition.
Technical Specifications
BookRead API exposes a standard cloud interface for synthetic voice synthesis. Send a POST request to the speech endpoint with your text payload to receive an audio file. The system processes requests sequentially, ensuring consistent output quality without requiring complex infrastructure management on your end.
Integration is straightforward using standard HTTP methods. The API accepts JSON bodies containing model selection, voice presets, and audio format preferences. This design allows developers to incorporate audio generation directly into automated publishing pipelines or manual production workflows with minimal code overhead.
API endpoint integration
The primary endpoint is POST https://api.regal33.info/v1/audio/speech. Authentication is handled via a Bearer token in the Authorization header. Requests are processed in strict sequence, meaning throughput depends on queue availability during high load periods.
Each request accepts up to 4096 characters of input text. For longer manuscripts, clients must split the text, generate individual audio segments, and join them client-side. Failed synthesis requests incur no charge, ensuring billing accuracy for production environments.
- Endpoint: POST https://api.regal33.info/v1/audio/speech
- Auth: Authorization: Bearer <key>
- Max input: 4096 characters per request
OpenAI SDK compatibility
The request structure mirrors the OpenAI speech endpoint exactly. By updating the base URL to https://api.regal33.info/v1, existing OpenAI SDK implementations work without modification. This reduces integration time for teams already familiar with standard TTS patterns.
Use the standard SDK methods to pass text, voice, and format parameters. The API supports both tts-1 and tts-1-hd models, allowing you to switch quality levels via a single parameter change while maintaining the same code structure.
- Base URL: https://api.regal33.info/v1
- Compatible with official OpenAI SDKs
- Identical request shape to OpenAI speech
Supported audio formats
Output formats include mp3, wav, opus, flac, and pcm. Select the format based on your distribution needs; mp3 and opus are suitable for web distribution, while wav and flac provide lossless quality for archival purposes. PCM is available for raw audio processing.
All generated files contain an inaudible synthetic watermark to identify the content as AI-generated. This feature supports transparency requirements for audiobook platforms and helps distinguish synthetic narration from human-recorded tracks in your catalog.
- mp3, wav, opus, flac, pcm
- Inaudible synthetic watermark included
- No streaming or WebSocket support
Pricing and Usage
BookRead API charges $0.01 per 1,000 characters, including spaces and punctuation. Failed requests incur no charge, with the full amount refunded. There are no monthly plans, seats, or minimum spend requirements, allowing you to scale production volume precisely to your needs without infrastructure overhead.
Pay-per-character model
The rate is $10 per 1 million characters processed. Each API request supports up to 4,096 characters of input text. For longer manuscripts, split the text into parts and join the resulting audio files to maintain consistent quality and manageable request sizes.
Trial credit availability
Signing in with Google provides $0.50 in trial credit, valid for 7 days, covering approximately 50,000 characters. Accounts created via email and password do not receive this initial credit. The balance does not expire once funded.
Top-ups are processed via USDT on the TRON network (TRC-20) with a $5 minimum. Credits are credited automatically within minutes of transfer confirmation.
- Trial: $0.50 for Google sign-ups (7 days)
- Minimum top-up: $5 via USDT (TRC-20)
- Balance never expires
Transparent billing cycles
We do not issue invoices with VAT or offer subscriptions. Request metadata, including time, character count, and status, is retained solely for billing purposes. Input text is used only for generation and is not stored or used for training.
Audio files are returned in the response and are not stored on our servers. You maintain full control over your generated assets without long-term storage fees.

Language Support
BookRead API supports 23 languages, enabling audiobook producers to reach global markets without hiring multiple voice actors. The system handles Arabic, Chinese, European, and Asian languages with consistent synthetic quality, ensuring uniform production standards across diverse titles.
Accurate language detection reduces manual configuration, but explicit setting remains necessary for ambiguous scripts. This approach minimizes errors in Latin-script texts that do not default to English, streamlining the workflow for multilingual self-publishing projects.
23 supported languages
The API covers a comprehensive range of markets, 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 preset can speak any of these languages, allowing flexible casting regardless of the voice's native origin.
Automatic language detection
If the language field is omitted, the system guesses the language from the text script. Latin-script text defaults to English, which is efficient for English-dominant catalogs but requires attention for mixed-language content.
Explicit language setting
For non-English Latin-script text or mixed scripts, explicitly set the language parameter using the ISO code. This ensures correct pronunciation and prosody. Note that Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, and Bengali are not supported.
Data Privacy and Rights
BookRead API processes input text exclusively for the purpose of generating the requested audio file. Once synthesis is complete, the generated audio is returned directly in the response and is not stored on our servers. This ensures that your manuscript content remains transient during processing.
Input text usage policy
Your text is used solely to produce the audio output for the specific request. We do not retain the full text content after the audio file is delivered, ensuring your narrative assets remain private to your workflow.
Synthetic watermarking
Every audio file generated by BookRead API contains an inaudible watermark. This marks the file as synthetic speech, providing clear provenance for your audiobook productions without affecting audio quality or listener experience.
No training data retention
Unlike some consumer-grade services, we do not use your input text to train our models. Your manuscripts are not added to our training dataset, preserving the uniqueness of your published works.
- Request metadata (time, characters, status) is retained strictly for billing purposes.
- No personal data from your text is used for model improvement.
Integration Guide
Integration is straightforward for audiobook producers using standard HTTP libraries. The API follows the OpenAI request shape, allowing you to switch base URLs without changing your code structure. This compatibility ensures that existing workflows for text-to-speech generation remain stable and predictable.
Quick start with cURL
Send a POST request to https://api.regal33.info/v1/audio/speech with your JSON body. Include your Authorization: Bearer key, the model identifier (tts-1 or tts-1-hd), and the text input. The response returns the audio file directly in the specified format, such as mp3 or wav.
SDK configuration
Use the official OpenAI SDKs by updating the base_url to https://api.regal33.info/v1. Pass your API key as the authentication token. This method works seamlessly with existing Python or Node.js scripts, requiring only the endpoint adjustment to route requests to BookRead API.
Error handling basics
Failed requests do not incur charges; the API returns a standard error code and the charge is returned in full. Monitor the response status to catch exceptions during synthesis. Since requests are processed sequentially, a high volume of concurrent calls may experience queue delays, which should be handled with appropriate retry logic.
Why Choose BookRead API
BookRead API delivers a straightforward integration experience designed for audiobook producers. Our endpoint mirrors the OpenAI speech structure, allowing you to use existing SDKs with a simple base URL change. This reduces development time and eliminates the need for custom wrapper libraries.
Pricing is strictly per character at $10 per million, with no hidden fees or monthly subscriptions. Failed requests are refunded automatically, ensuring you pay only for successful audio generation. This model supports scaling without infrastructure overhead.
Simplicity of the API
Integration requires only a standard HTTP POST request. You send text and receive an audio file without managing complex streaming pipelines or WebSocket connections. The consistent request shape allows for easy batching of up to 4096 characters per call.
- Standard JSON body structure
- Compatible with OpenAI SDKs
- No custom protocol required
Predictable costs
Costs are calculated by character count, including spaces and punctuation. At $0.01 per 1,000 characters, expenses remain consistent regardless of audio length. There are no tiered pricing models or volume discounts to negotiate.
No enterprise lock-in
We do not require minimum spend or long-term contracts. You can issue a new API key at any time, instantly invalidating the old one. There are no proprietary SDKs or vendor-specific dependencies locking you into our infrastructure.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. A mismatch between the voice’s inherent tone and the listener’s age or preference can reduce engagement, regardless of audio fidelity.
23 languages
Any voice speaks any of these languages.
