Frequently asked questions
Which voice is best for history tours?
Orion has a deep, documentary style. Atlas provides a neutral, news-like tone suitable for factual presentations. Luna offers a soft, storytelling approach for immersive historical narratives. You can test all voices to see which best matches your specific tour content and audience expectations.
Can I use a male voice for a female audience?
Yes. Any of the eight voices can be used for any audience. There is no restriction based on listener gender. You can choose Milos friendly conversational tone or Atlass neutral news style regardless of whether your visitors are male or female. The voice choice depends on your content style, not the audience demographics.
Do voices sound different in different languages?
The voice presets remain consistent, but the system guesses the language from the script. If you provide non-English Latin text without setting the language field, it defaults to English pronunciation. For accurate results in other languages, always specify the ISO language code. This ensures correct pronunciation across all 23 supported languages.
Is there a difference between tts-1 and tts-1-hd voices?
No. Both models use the same engine and have identical pricing. The only difference is the model identifier in your API request. You can use either "tts-1" or "tts-1-hd" to generate audio. The output quality and voice characteristics are the same for both options.
Can I mix voices in a single tour?
Yes. You can generate separate audio files for different sections using different voices. For example, use Luna for the introduction and Atlas for factual exhibits. Since the API processes requests sequentially, you can combine these files into a single tour guide. This allows for varied narration styles within one experience.
Are the voices robotic or natural?
The voices are natural, based on licensed recordings from LibriTTS-R and LibriVox. They are not robotic. Every generated file includes an inaudible watermark marking it as synthetic speech. The tone is practical and clear, suitable for tourism and museum contexts without sounding like standard text-to-speech robots.
Do voices have accents?
The voices do not have strong regional accents by default. They speak clearly in any of the 23 supported languages. If you do not specify a language, Latin script defaults to English. To avoid unintended accents, always set the language parameter explicitly when using non-English text for your audio guides.
Can I adjust the pitch of the voices?
No. You can adjust the speed between 0.25 and 4.0, but pitch remains fixed. There is no SSML support for pitch control. You cannot modify the fundamental frequency of the voices. To change the perceived tone, select a different voice preset that matches your desired vocal range and style.
How do I choose between Aria and Luna?
Choose Aria for warm, clear narration suitable for general information. Choose Luna for a soft, storytelling approach ideal for immersive historical or artistic content. Aria is better for straightforward guides, while Luna enhances narrative flow. Test both voices with your script to determine which best fits your museum or tour atmosphere.
Is Atlas good for news updates?
Yes. Atlas is designed with a neutral, news-like tone. It is ideal for delivering factual updates, schedules, or informational announcements in museums. Its clear, professional sound ensures that information is communicated effectively without emotional bias, making it a reliable choice for standard informational content in tourism settings.
Can I use Felix for a children's tour?
Yes. Felix has an energetic, promo-style tone that can engage younger audiences. While not specifically designed for children, its upbeat nature works well for dynamic tours. You can adjust the speed to suit the pacing of a childrens tour. Test Felix with your script to ensure it matches your desired energy level.
Do the voices work well in noisy museums?
The voices are clear and natural, but audio quality depends on your playback hardware. The API outputs standard formats like mp3, wav, or opus. In noisy environments, higher volume or directional speakers help. The voices themselves do not have noise cancellation.
Are there any voice limitations?
You are limited to eight specific voice presets. There is no custom voice cloning or emotion tagging. You cannot upload your own voice. Each request supports up to 4096 characters. Voices are based on licensed recordings and carry an inaudible synthetic watermark. These are the only constraints on voice usage.
How do I test voices before buying?
Sign up with Google to receive $0.50 of free trial credit. This allows about 50,000 characters of generation. Use the API to generate audio with different voices. Check your balance and usage on the dashboard. This trial lets you evaluate voice quality and speed before committing to paid top-ups via USDT.
Female Voices for Narration
Aria, Nova, Luna, and Sage offer distinct tones for tourism content. Aria provides a warm, clear narration ideal for historical sites. Nova delivers bright, upbeat energy for modern exhibits. Luna creates soft, storytelling atmospheres for relaxation-focused tours. Sage ensures calm, instructional clarity for museum guides.
Aria: Warm and clear for history
Aria’s voice suits historical narratives, delivering clarity without sounding artificial. Her tone respects the gravity of heritage sites while remaining engaging for visitors.
- Ideal for chronological historical tours
- Clear enunciation for complex dates and names
- Warm tone enhances emotional connection
Nova: Bright and upbeat for modern exhibits
Nova brings energy to contemporary exhibits, keeping visitors engaged. Her upbeat tone works well for interactive displays or modern art descriptions.
Use Nova when you want to highlight innovation or contemporary themes in your museum.
- High energy for dynamic exhibits
- Engaging for younger audiences
- Clear articulation for detailed descriptions
Luna: Soft and storytelling for relaxation
Luna’s soft voice creates a calming atmosphere, perfect for meditation spaces or nature trails. Her storytelling style draws listeners in without distraction.
- Soft delivery for tranquil environments
- Narrative style for immersive experiences
- Reduced auditory fatigue for long tours
Sage: Calm and instructional for museums
Sage offers a calm, instructional tone ideal for detailed museum guides. Her consistent pacing helps visitors absorb information without feeling rushed.
- Consistent pace for complex information
- Calm tone reduces cognitive load
- Professional delivery for educational content
Male Voices for Information
Felix, Milo, Atlas, and Orion provide versatile male voices for museum and tourism audio guides. Each voice is crafted from licensed recordings, ensuring clear and professional narration for diverse content types.
Felix: Energetic for promotional tours
Felix brings high energy to promotional content and dynamic city tours. His tone is engaging and keeps listeners interested during fast-paced audio segments.
- Ideal for short promotional clips
- High energy delivery
- Best for urban or action-oriented tours
Milo: Friendly for interactive kiosks
Milo offers a warm, conversational tone perfect for interactive kiosks. His friendly demeanor makes visitors feel welcomed and guides them naturally through exhibit information.
- Conversational and approachable
- Optimized for kiosk audio
- Supports 23 languages
Atlas: Neutral for news and updates
Atlas delivers a clear, neutral voice ideal for factual updates and news. His consistent tone ensures information is delivered without emotional bias, suitable for time-sensitive announcements.
- Neutral and clear
- Best for news and updates
- Consistent tone
Orion: Deep for documentary-style tours
Orion provides a deep, authoritative voice for documentary-style tours. His rich tone adds depth and seriousness to historical or educational content, enhancing the listening experience.
- Deep and authoritative
- Ideal for documentaries
- Enhances educational content
Matching Voices to Tour Types
Selecting the right voice ensures your audio guide matches the visitor experience. By aligning vocal tone with tour content, you enhance engagement and retention without requiring complex production steps.
History tours: Aria or Orion
Aria’s warm, clear narration suits historical narratives, providing a consistent and engaging tone for heritage sites. Orion’s deep, documentary style adds gravity to historical events, ideal for serious historical contexts.
- Aria: Warm and clear for storytelling history
- Orion: Ideal for documentary-style history
Nature walks: Luna or Milo
Luna offers a soft, storytelling voice perfect for relaxation-focused nature walks. Milo provides a friendly, conversational tone that works well for interactive kiosk experiences in parks.
- Luna: Soft and storytelling
- Milo: Friendly and conversational
City guides: Felix or Nova
Felix delivers an energetic, promo-style voice suitable for vibrant city tours. Nova’s bright, upbeat tone keeps energy high for modern urban exploration and shopping district guides.
- Felix: Energetic and promo-style
- Nova: Bright and upbeat
Voice Consistency Across Languages
Create consistent audio guides across multiple locations without re-recording. Our API supports 23 languages, allowing you to maintain a uniform listener experience whether your tour is in English, Spanish, or Japanese.
The same voice preset speaks all supported languages naturally. You do not need to swap voice profiles when translating scripts, ensuring brand consistency across every language variant.
Keep your tour’s identity cohesive. Use the same voice preset for all language versions to build recognition and trust with visitors, regardless of their native tongue.
All voices support 23 languages
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.
Any voice can speak any of these languages. If you do not set the language parameter, the system guesses the language from the script (defaulting to English for Latin script).
- Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish
No need to change voices for translation
Use the same voice preset across all language versions of your script. This maintains a consistent auditory identity for your museum or tourism brand.
Simply pass the language parameter in your request to ensure correct pronunciation. The engine handles the linguistic differences without altering the voice character.
Maintain brand identity in audio
Consistency builds familiarity. Visitors recognize your guide by voice, even when the language changes. This uniformity strengthens brand recall and professional presentation.
Select one voice preset for your entire tour series. Whether a visitor is in Tokyo or Paris, the tone and character remain identical.
Audio Quality and Format
Audio clarity is critical for museum speakers and outdoor tourism environments where background noise is common. Our API generates clean audio files that ensure every word is intelligible, whether played through large hall speakers or small handheld devices.
High clarity for museum speakers
The synthesis engine produces clear, natural-sounding speech optimized for playback on standard audio systems. This ensures your tour scripts are easy to understand even in acoustically challenging spaces like grand halls or busy plazas.
MP3 for web streaming
MP3 is a common format for web-based audio guides. It offers a small file size that loads quickly on mobile devices, ensuring smooth playback for tourists accessing content via their smartphones or web kiosks.
WAV for high-fidelity playback
Choose WAV when you need uncompressed, high-fidelity audio for dedicated hardware or professional installations. This format preserves the full quality of the voice, making it suitable for high-end museum exhibits where audio fidelity is a priority.
Testing Voice Options
Testing voice options ensures your audio guide resonates with visitors before full production. VoxVoyage provides a free trial with $0.50 credit (about 50,000 characters) valid for 7 days upon Google sign-in, allowing you to evaluate multiple voices without financial risk.
Use this trial period to generate short samples of your actual tour script. Compare how each preset interprets your specific content, noting clarity, tone, and natural flow. This practical approach helps you select the most suitable voice for your audience.
Use the free trial to test voices
Sign in via Google to receive $0.50 in trial credit, valid for 7 days. This allows immediate testing without payment setup. Use the API to generate audio from your actual tour script snippets, comparing different presets directly.
The free trial gives you about 50,000 characters to experiment. You can test multiple voices on the same text to hear subtle differences in warmth, energy, and clarity. This low-risk method helps you identify the best fit for your specific content.
- Sign up with Google for instant trial credit
- Use 50,000 characters to test multiple voices
- No payment method required to start
Generate sample clips for each
Generate short audio clips for each of the eight presets using the same 100–200 character text segment. This ensures a fair comparison. Listen to how each voice handles your specific wording, punctuation, and pacing.
You can test voices in different languages if your guide supports multiple tours. Remember that any voice can speak any of the 23 supported languages, so you don't need separate voices for translations. Focus on consistency and clarity.
- Test identical text across all 8 voices
- Evaluate handling of punctuation and pacing
- Compare performance in multiple supported languages
Choose based on listener feedback
Share your generated samples with a small group of test listeners or staff. Ask for feedback on clarity, tone, and engagement. Different environments may suit different voices; a busy museum lobby might need the clear, calm tone of Aria, while a nature walk might benefit from Luna's storytelling quality.
Once you select the best voice, proceed to generate the full audio library. Keep your test samples for reference during quality assurance. This feedback loop ensures your final audio guide meets visitor expectations without unnecessary rework.
- Gather feedback from test listeners or staff
- Match voice to specific tour environment
- Keep samples for quality assurance reference
Technical Requirements
Generate audio files via our API and play them on any standard device. No proprietary hardware or specialized players are required to deliver your tourism content.
No special hardware needed
Your existing infrastructure works seamlessly. Connect the API to your website or kiosk software without buying new equipment.
- Compatible with standard PCs and mobile devices
- Works with standard web players
- Supports standard audio formats
Works with standard web players
Audio guides play in any modern browser. Embed the generated MP3 or WAV files directly into your existing site structure.
Supports standard audio formats
Choose between MP3 for efficient web streaming or WAV for high-fidelity playback. Both formats ensure compatibility with common audio hardware.

Updating Your Audio Library
Regenerate audio with new voices to refresh your collection without losing existing files. Since the API returns audio directly in the response, you can swap voices for any script instantly. This allows you to test different narrators for the same tour content without re-recording from scratch.
Regenerate audio with new voices
You can generate new audio files using any of the eight preset voices. Simply send the same text with a different voice parameter. The old files remain untouched, allowing you to compare performance.
Keep old files for comparison
Store previous audio versions alongside new generations. This helps you evaluate which voice best suits your museum or tour environment. You can listen side-by-side to decide if a new voice offers better clarity or tone for your visitors.
Maintain version control
Label your audio files with version numbers or dates. This ensures you know which voice model generated each clip. Consistent naming helps you manage updates across large audio libraries without confusion.
- Use descriptive filenames like
tour_history_v2.mp3 - Keep a log of voice choices per script
- Archive unused voices to save storage space
Accessibility Considerations
Accessibility is critical for museum and tourism audio guides. Our API ensures clear pronunciation across 23 languages, supporting diverse visitor audiences without requiring specialized hardware or complex setups.
Clear pronunciation for diverse audiences
Voices like Aria and Atlas deliver precise articulation, ensuring historical facts and directions are understood by non-native speakers and tourists with varying language backgrounds.
- Supports 23 languages including Arabic, Chinese, Hindi, and Japanese
Adjustable speed for hearing impaired
You can adjust speech speed from 0.25x to 4.0x. This allows visitors with hearing impairments to slow down audio for better comprehension or speed it up for quick updates.
- Adjustable speed range: 0.25–4.0
Consistent tone for cognitive ease
Using a single voice preset across all guide content maintains a consistent tone. This reduces cognitive load for visitors, making long tours easier to follow and less fatiguing.
Cost vs. Quality Trade-offs
We offer two model options for your audio guides: tts-1 for standard clarity and tts-1-hd for enhanced naturalness. Both models deliver high-quality synthetic speech suitable for museum speakers and tourism apps, ensuring your content sounds professional and engaging.
Standard vs. HD quality options
The tts-1 model provides clear, natural-sounding speech optimized for general use. The tts-1-hd model uses a different engine but has the same price.
Same price for both models
Unlike many services, we charge $0.01 per 1,000 characters regardless of the model selected. Whether you choose the standard or HD engine, your cost remains identical, allowing you to upgrade audio quality without increasing your budget.
Choose based on audio environment
Select tts-1-hd if you prefer the HD engine. Use tts-1 for standard guides.
Voice Licensing and Rights
VoxVoyage voices are synthesized from licensed recordings, ensuring your audio guides are built on a solid legal foundation. This background licensing covers the source material used to train the synthetic voices, allowing you to use the generated audio commercially without worrying about underlying rights disputes.
Voices based on licensed recordings
The underlying recordings come from the LibriTTS-R dataset, which is licensed under CC BY 4.0. This ensures the foundational audio data is legally cleared for commercial use.
- Source data includes LibriTTS-R (CC BY 4.0)
- Includes LibriVox public-domain readings
Clear usage rights for tours
You own the generated audio files. You can play them in museums, on websites, or in tour apps without seeking further permission. The synthetic nature of the output means you are licensing the result, not the underlying voice talent's ongoing performance rights.
No additional licensing fees
There are no per-play fees or royalties to pay for your audio guides. Once you have generated the file, you can distribute it indefinitely. The only cost is the character usage during generation, with no hidden licensing charges for the output itself.
Future-Proofing Your Audio
Long-term audio projects require stable infrastructure. VoxVoyage uses a standard API format that ensures your audio generation workflows remain compatible with evolving web standards. This approach prevents vendor lock-in and keeps your archive accessible for years.
By relying on widely adopted protocols, you avoid proprietary constraints that might limit future updates or integrations with new museum display systems.
Standard API format for longevity
Our endpoint structure mirrors industry standards, ensuring that your integration code remains relevant as technology shifts. This stability means less maintenance overhead for your team.
Audio files generated today will remain playable in future software environments without requiring re-encoding or format conversion.
- JSON-based request bodies ensure consistent data handling.
- Predictable endpoint URLs reduce integration errors.
- Support for common audio formats like MP3 and WAV ensures broad compatibility.
Open SDK compatibility
You can use standard OpenAI SDKs by simply adjusting the base URL. This flexibility allows you to switch or upgrade client libraries without rewriting your core logic.
This open approach means you are not tied to a single vendor’s specific implementation, giving you freedom to choose the best tools for your needs.
- Compatible with official OpenAI SDKs.
- Easy to integrate with existing Python or Node.js projects.
- Standard HTTP requests simplify debugging and monitoring.
Easy migration if needed
If your requirements change, you can easily switch voices or update scripts without losing your existing audio assets. Your generated files are standalone.
The simple character-based billing and clear API structure make it straightforward to scale usage up or down as your museum’s audience grows.
Migrating to a different provider or upgrading your internal systems is simplified because the input and output formats are standardized.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. This feedback loop ensures your final audio guide meets visitor expectations without unnecessary rework.
23 languages
Any voice speaks any of these languages.
Sources and further reading
- Wikipedia
- curl documentation
- OpenAI API reference: create speech
- Node.js documentation
- openai-python on GitHub
- LibriVox public-domain audiobooks
- openai-node on GitHub
- Creative Commons Attribution 4.0
- W3C WAI: making audio and video accessible
- ISO 639 language codes
- Tether (USDT)
- LibriTTS-R paper (arXiv 2305.18802)
