Choosing the Right Voice for Kindergarten Audiobooks

By Lumivox editorial team · Updated 2026-11-18

This guide helps self-publishers and kindergarten content creators select the right text-to-speech voice for their audiobooks. We list eight preset voices—four female and four male—describing their distinct tones, from warm narrators to energetic storytellers. You will learn how to match voice styles to your story’s mood and audience age.

We also cover technical details relevant to long-form content, including language support, character limits, and pricing. Since all voices share the same engine and cost, your choice depends entirely on narrative fit. Read on to find a suitable voice for your next children’s book.

Choosing the Right Voice for Kindergarten Audiobooks

Frequently asked questions

Which voice is best for bedtime stories?

Luna is described as soft and storytelling, making it suitable for bedtime stories. Aria is a warm, clear narrator suitable for general reading. You can choose between these female presets based on your preferred tone. Each voice offers a distinct character to match the mood of your kindergarten audiobook without requiring additional settings.

Can I use a male voice for a girl's story?

Yes, any of the eight voices can speak any of the 23 supported languages. You are not restricted by gender. For example, you could use Felix, Milo, Atlas, or Orion for a girl’s story. The voice selection is independent of the character’s gender in your text.

Do voices sound different in different languages?

Each voice has a specific timbre that remains consistent across languages. However, the pronunciation and intonation will adapt to the language spoken. If you do not set the language parameter, the system guesses it, defaulting to English for Latin script. Setting the language ensures accurate pronunciation.

Is there a voice for educational content?

Sage is described as calm and instructional, making it well-suited for educational content. Atlas is neutral and news-like, also appropriate for clear instruction. You can adjust the speed between 0.25 and 4.0 to control the pacing for your audience’s comprehension needs.

Can I combine voices in one book?

The API generates audio for one voice per request. To combine voices, you must generate separate audio files for each section or character using the desired voice preset. Then, you join these audio files manually or via your own pipeline. The API does not support multi-voice tracks in a single request.

What is the difference between tts-1 and tts-1-hd?

Both models use the same engine and have the same price. The only difference is the model identifier in your request. You can use either 'tts-1' or 'tts-1-hd' without any change in cost or functionality. They produce identical audio quality for your kindergarten audiobooks.

Are the voices based on real people?

Yes, the voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain audiobook readings. They are synthetic recreations, not direct recordings of specific individuals. Every generated file carries an inaudible watermark that marks it as synthetic speech.

Can I clone a voice for my book?

No, custom voice cloning is not offered. You cannot upload your own voice or design a new one. You must choose from the eight available presets: Aria, Nova, Luna, Sage, Felix, Milo, Atlas, or Orion. These are the only voices available for your audiobooks.

Do voices have accents?

The voices are designed to speak clearly in their supported languages. Accents are not explicitly customizable. The system guesses the language from the script if not specified, defaulting to English for Latin text. Setting the language parameter ensures the voice uses the correct pronunciation for that language.

How do I listen to voice samples?

You can listen to voice samples by calling the GET /v1/voices endpoint. This endpoint returns information about voices, languages, formats, and price. It provides the necessary details to select the best voice for your kindergarten audiobook before you generate any audio.

Is the audio quality consistent across voices?

All voices use the same engine and pricing model. The audio quality is consistent in terms of format and clarity. Differences lie only in the vocal characteristics of each preset. You can generate audio in mp3, wav, opus, flac, or pcm formats.

Can I change the voice mid-book?

Yes, by generating separate audio files for different sections. You can switch voices between requests. Since the API handles long texts by splitting them, you can assign different voices to different parts of your book and join the resulting audio files.

Do voices support emotional expression?

No, emotion tags are not supported. Each voice has a fixed tone: Aria is warm, Luna is soft, Felix is energetic, etc. You cannot add emotional variations like happiness or sadness dynamically. The emotional quality is determined by the voice preset you select.

What is the watermark in the audio?

Every generated file carries an inaudible watermark that marks it as synthetic speech. This watermark identifies the audio as AI-generated. It is embedded in the audio file and is not audible to the listener. It serves as a marker of origin for your content.

Voice Overview

Sound wave illustration for: Voice Overview

Lumivox provides eight preset voices designed to handle the diverse needs of kindergarten audiobooks. Each voice is generated from licensed recordings, ensuring a consistent and reliable output for your self-published stories. The API handles the technical heavy lifting, allowing you to focus on the narrative.

The system offers a balanced mix of male and female options, each with distinct tonal qualities. Whether you need a warm narrator or a bright, energetic guide, you can select the voice that matches your story’s mood. All voices support 23 languages, ensuring broad accessibility for your audience.

8 preset voices available

Choose from eight distinct voice presets, each crafted for specific storytelling styles. The library includes both female and male voices, giving you the flexibility to match the tone of your kindergarten book. From warm narrators to energetic guides, every voice is designed for clarity and engagement.

  • Aria, Nova, Luna, Sage (Female)
  • Felix, Milo, Atlas, Orion (Male)

Male and female options

The female voices offer a range of tones, from the warm and clear narration of Aria to the soft, storytelling quality of Luna. Nova brings a bright, upbeat energy, while Sage provides a calm, instructional tone. These options allow you to create a comforting and engaging listening experience for young readers.

Different tones and styles

Male voices include Felix, who delivers an energetic, promo-style delivery, and Milo, who offers a friendly, conversational tone. Atlas provides a neutral, news-like clarity, while Orion offers a deep, documentary feel. This variety ensures you can find a suitable voice to bring your kindergarten book to life.

Female Voices

Sound wave illustration for: Female Voices

Female voices offer distinct emotional ranges for children’s narratives, from gentle narration to energetic storytelling. Aria, Nova, and Luna each provide a unique tone suitable for different story types, ensuring your kindergarten audiobooks capture the right mood.

These presets are based on licensed recordings and work across all supported languages. Selecting the right voice helps maintain listener engagement while keeping production simple and consistent.

Aria: warm and clear

Aria delivers a warm, clear narrator style suitable for bedtime stories or gentle educational content. Her tone remains consistent even over longer texts, making her a reliable choice for series books where listeners hear the same voice repeatedly.

  • Ideal for calm, narrative-driven stories
  • Clear articulation aids young listeners
  • Consistent tone across long texts

Nova: bright and upbeat

Nova brings a bright, upbeat energy perfect for interactive stories or songs. Her lively delivery keeps children engaged, especially for shorter, action-oriented tales where pacing and enthusiasm drive the experience.

  • High energy for interactive content
  • Engaging for short stories
  • Upbeat tone maintains attention

Luna: soft and storytelling

Luna offers a soft, storytelling voice that feels intimate and soothing. She excels in fantasy or emotional narratives, providing a gentle backdrop that allows young listeners to immerse themselves in the plot without distraction.

  • Soft tone for emotional stories
  • Great for fantasy or fairy tales
  • Immersive and soothing delivery

Male Voices

Sound wave illustration for: Male Voices

Male voices offer distinct tonal qualities for different narrative needs. Felix delivers high-energy delivery ideal for promotional segments or upbeat reading. Milo provides a warm, conversational tone that feels like a friend reading aloud. Atlas maintains a neutral, clear style suitable for structured content or news-style storytelling within children’s books.

Felix: energetic and promo

Felix excels at maintaining high energy levels, making him suitable for promotional segments or lively story introductions. His voice keeps young listeners engaged with a dynamic tone that avoids monotony.

  • Ideal for upbeat story openings
  • Maintains listener attention through varied pacing

Milo: friendly and conversational

Milo offers a friendly, conversational tone suitable for reading a bedtime story. His approachable tone reduces the distance between narrator and listener, creating a comfortable atmosphere for self-published kindergarten audiobooks.

  • Best for intimate, one-on-one reading styles
  • Reduces listener fatigue during long sessions

Atlas: neutral and news

Atlas offers a balanced, neutral delivery that ensures clarity without emotional bias. This makes it perfect for educational segments or straightforward narratives where the focus remains on the story facts rather than dramatic interpretation.

  • Clear articulation for educational content
  • Consistent tone across diverse story types

Voice Selection Criteria

Sound wave illustration for: Voice Selection Criteria

Selecting a voice is the first step in producing a kindergarten audiobook that keeps young listeners engaged. The right tone transforms static text into an immersive experience, making the story feel alive rather than just read aloud.

Match story mood

Align the voice’s inherent character with your book’s narrative arc. Use Luna’s soft, storytelling quality for gentle bedtime tales, while Aria’s clear narration suits educational content. Match Felix’s energy to adventure segments or Milo’s friendliness for conversational stories.

  • Use Luna for gentle, bedtime stories.
  • Choose Aria for clear, educational content.
  • Select Felix for high-energy adventures.
  • Pick Milo for conversational narratives.

Consider listener age

Younger children respond better to brighter, higher-pitched tones like Nova or Luna. Older kids or parents reading to toddlers may prefer the calm, instructional tone of Sage or the neutral clarity of Atlas, which remains engaging without being overly childish.

  • Nova and Luna suit younger listeners.
  • Sage and Atlas work for older kids.
  • Avoid overly dramatic tones for toddlers.

Test clarity and tone

Generate short samples using different voices to test pronunciation and pacing. Ensure the voice remains clear at various speeds (0.25–4.0) and that the audio quality meets your standards before committing to a full book. Listen for consistency across chapters.

  • Generate samples for each voice.
  • Test different speed settings.
  • Check consistency across chapters.

Technical Constraints

Sound wave illustration for: Technical Constraints

This guide outlines the technical boundaries of the Lumivox API for kindergarten audiobook production. Understanding these constraints ensures smooth integration and realistic expectations for your audio files.

All voices support 23 languages

Any of the eight voices can speak any of the 23 supported languages, including English, Spanish, and Chinese. If the language field is omitted, the system guesses the language based on the script, defaulting to English for Latin text. For non-English Latin-script texts, explicitly set the language code to ensure accuracy.

No custom voice cloning

The service offers eight fixed presets rather than custom voice cloning. You cannot upload your own voice or create unique variations. Each generated file is based on licensed recordings from LibriTTS-R and LibriVox public-domain audiobooks.

Watermarked synthetic speech

Every audio file produced contains an inaudible watermark identifying it as synthetic speech. This metadata is embedded automatically and cannot be removed. It serves to mark the content as AI-generated without affecting the listening experience.

Performance Notes

Sound wave illustration for: Performance Notes

Generating kindergarten audiobooks requires handling texts that often exceed the 4,096-character limit. The API automatically splits input into manageable parts, synthesizes each segment, and joins the audio files into a single continuous file. This ensures your long stories are processed automatically.

For texts longer than a few seconds, generation speed is roughly real-time, averaging about 25 characters of text per second of audio. Short phrases may take a few seconds to process, but long narratives progress at a steady pace proportional to their length.

Real-time speed on long texts

The engine processes text at approximately real-time speed for longer content. You can expect about 25 characters of text to generate one second of audio. This predictable pace helps you estimate production times for your kindergarten book projects accurately.

  • Approximately 25 characters of text per second of audio
  • Automated splitting for texts over 4,096 characters
  • Automatic joining of segmented audio files

Queue under load

Requests are processed sequentially. If many users are generating audio simultaneously, your request may wait in a queue. During high traffic, generation times may increase slightly as requests are handled one after another.

  • Requests processed sequentially
  • Potential delays during high traffic
  • No priority tiers for standard keys

Consistent quality

Each segment maintains the same voice and tone as the previous one. The joining process creates a continuous file for playback. Quality remains uniform across all generated segments, regardless of text length.

  • Uniform voice characteristics across segments
  • Seamless audio joining
  • Consistent tone throughout the narrative

Licensing Info

Sound wave illustration for: Licensing Info

Lumivox voices are derived from the LibriTTS-R dataset, which utilizes recordings from LibriVox. These are public domain audiobook readings, ensuring that the core vocal textures are free from restrictive copyright claims. This foundation allows us to offer synthetic voices based on licensed recordings.

Based on LibriTTS-R

The underlying vocal models are trained on the LibriTTS-R dataset. This dataset provides a robust variety of natural speech patterns, which we refine to create our eight distinct voice presets. The data comes from volunteer readers narrating public domain books, giving us a rich source of clear, articulate speech suitable for children's content.

  • Trained on LibriTTS-R dataset
  • Derived from LibriVox recordings

CC BY 4.0 licensed

The source audio data is licensed under Creative Commons Attribution 4.0 (CC BY 4.0). This means the original recordings are freely available for commercial and non-commercial use, provided attribution is given to the original readers and platforms. This license structure supports our model of offering affordable, unrestricted TTS services.

Public domain sources

Most content in the source dataset consists of public domain texts, such as classic fairy tales and stories where copyright has expired. This ensures that the narrative content itself is not encumbered by third-party rights. You can generate audio for these stories without worrying about underlying text copyright conflicts.

Licensing Info

Integration Tips

Sound wave illustration for: Integration Tips

Integrating Lumivox into your kindergarten audiobook workflow is straightforward if you leverage existing tools. The API mirrors the OpenAI speech endpoint structure, meaning you can use the official OpenAI SDKs with minimal code changes. This allows you to generate audio files directly within your preferred programming environment without learning a new library from scratch.

The primary adjustment involves pointing your SDK client to the correct base URL. Once configured, you can send text and receive audio files just as you would with any standard TTS service. This approach keeps your integration simple and maintainable for long-term self-publishing projects.

Use OpenAI SDK

The API request shape matches the OpenAI speech endpoint exactly. You can use the official OpenAI SDKs by simply updating the base URL configuration. This allows you to use familiar methods like client.audio.speech.create() without rewriting your entire integration logic.

This compatibility means you can integrate TTS generation into your existing Python or Node.js scripts with minimal effort. The SDK handles the JSON formatting and request headers for you, reducing the chance of syntax errors in your audiobook production pipeline.

  • Use the official OpenAI SDK for Python or Node.js
  • Keep familiar method calls like audio.speech.create
  • No need to learn a new library syntax

Set base URL correctly

Point your SDK client to https://api.getx99.com/v1 instead of the default OpenAI endpoint. This ensures that your requests are routed to the Lumivox infrastructure. All other parameters like model, voice, and input remain identical to the standard OpenAI format.

Verify that your API key is correctly set in the Authorization header. The key should start with tts- and be active for your account. Double-check the base URL to avoid confusion with other TTS services you might be using.

Handle errors gracefully

Implement basic error handling to catch network issues or invalid requests. If synthesis fails, the charge is returned in full, so you can retry without losing credit. Use try-catch blocks to manage exceptions during audio generation for your kindergarten stories.

Check the response status code to determine if the request was successful. If the API returns an error, log the details for debugging. This helps you maintain a smooth workflow when generating large batches of audiobook chapters.

Cost Considerations

Sound wave illustration for: Cost Considerations

Pricing for kindergarten audiobooks is straightforward: every voice costs the same. You pay $0.01 per 1,000 characters, regardless of whether you choose Aria for warm narration or Atlas for neutral storytelling. There are no tiered voice fees or premium voice upcharges.

The model selection does not affect your cost. Both tts-1 and tts-1-hd use the same engine and pricing structure. You can experiment with higher fidelity for complex nursery rhymes without incurring extra charges per voice or per style.

Same price for all voices

All eight preset voices—female and male—share identical pricing. Your budget depends solely on text length, not voice personality. This consistency simplifies cost estimation for self-publishers managing multiple book series.

  • No voice-specific surcharges
  • Uniform rate: $0.01 per 1,000 characters

tts-1 and tts-1-hd equal cost

Choosing tts-1-hd for clearer audio does not increase your bill. Both models charge the same rate per character. You can upgrade audio quality for educational content without changing your cost structure.

No extra fees for voices

There are no hidden fees for voice usage, style changes, or language switches within the supported set. Your balance decreases only by character count. Failed requests are free, ensuring you pay only for successful audio generation.

Language Support

Sound wave illustration for: Language Support

Supporting 23 languages, our API allows any voice to speak any listed language, giving you flexibility for diverse kindergarten stories. This means you can maintain a consistent character voice across different linguistic sections of a book.

Any voice speaks any language

All eight preset voices, from Aria to Orion, support the full range of 23 available languages. You are not restricted to specific voice-language pairs, allowing for creative storytelling approaches.

  • Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish.

Set language code for accuracy

For non-English Latin-script text, always set the language field. Without it, the system guesses the language, which may default to English and cause pronunciation errors for words like 'cake' or 'school' in other languages.

Auto-detect for Latin text

If you do not set the language, the API guesses based on the script. Latin text defaults to English, so ensure you specify the code for accurate pronunciation of non-English titles or phrases.

Common Mistakes

Sound wave illustration for: Common Mistakes

Many creators encounter avoidable errors when generating kindergarten audiobooks. Ignoring language codes or exceeding character limits disrupts the workflow, leading to unexpected results or failed synthesis requests.

Ignoring language code

If you submit Latin-script text without setting the language parameter, the API defaults to English. This causes mispronunciation for non-English stories. Always specify the ISO code for accurate delivery.

Exceeding character limit

Each request accepts up to 4,096 characters. Sending longer text causes the synthesis to fail. Ensure your input fits within this boundary before sending.

Not splitting long texts

For books exceeding the limit, split the text into parts of 4,096 characters or less. Generate audio for each part and join the files. This ensures consistent quality and avoids request failures.

Support Resources

Sound wave illustration for: Support Resources

For developers integrating the Lumivox API into their audiobook workflows, reliable support ensures smooth deployment. We provide direct email assistance for technical questions regarding voice selection, integration errors, or billing discrepancies. Our documentation is comprehensive, offering step-by-step guides for using the OpenAI SDK with our base URL, along with code examples for handling rate limits and audio generation.

Email support available

We offer email support for all technical and account-related inquiries. Unlike phone-based services, our team responds directly to your queries via email, ensuring detailed explanations for complex integration issues. There is no live chat or phone number; simply send your question to our support address.

  • Direct email access for technical and billing questions
  • No phone number or live chat available
  • Detailed responses for complex integration issues

Documentation linked

Our documentation covers all API endpoints, including voice parameters and request formats. It includes ready-to-use curl examples and SDK integration tips for Python and Node.js. Each section is structured to help you troubleshoot common errors, such as character limits or rate limit exceeds, without needing immediate human intervention.

  • Covers all API endpoints and request formats
  • Includes curl examples and SDK integration tips
  • Structured troubleshooting for common errors

FAQ section helpful

The FAQ section addresses frequent questions about voice presets, language support, and pricing. It clarifies that we do not offer custom voice cloning or SSML, helping you set realistic expectations. You can quickly find answers about character limits, audio formats, and trial credit usage before contacting support.

  • Answers questions about voice cloning and language support
  • Clarifies missing features like SSML and custom cloning
  • Quick access to pricing and trial credit details

Final Recommendations

Sound wave illustration for: Final Recommendations

Start by using the free trial credit to generate your first few chapters. This allows you to evaluate the quality without financial commitment before purchasing top-up credit.

Test multiple voices to find the best fit for your specific narrative style. Compare different tones to ensure the audio matches the mood of your kindergarten stories.

Start with trial credit

Use the $0.50 trial credit to generate initial audio samples. This credit covers approximately 50,000 characters, allowing you to test the API without immediate payment.

Test multiple voices

Generate short clips with Aria, Nova, Luna, Felix, Milo, Atlas, and Orion. Listen for clarity and engagement to determine which voice best suits your content.

Keep notes on which voices maintain listener interest for longer texts. Use these insights to streamline your production workflow.

Choose based on story tone

Match the voice to the story’s mood. Use warm narrators for gentle tales and energetic voices for upbeat adventures. Consistency in tone helps build a recognizable brand for your audiobooks.

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. Male voices offer distinct tonal qualities for different narrative needs.

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. LibriTTS-R corpus (OpenSLR 141)
  3. W3C WAI: making audio and video accessible
  4. openai-node on GitHub
  5. FFmpeg documentation
  6. ISO 639 language codes
  7. Python documentation
  8. openai-python on GitHub
  9. LibriTTS-R paper (arXiv 2305.18802)
  10. Tether (USDT)
  11. curl documentation
  12. LibriVox public-domain audiobooks

More on this site