Frequently asked questions
Which voice is best for navigation?
No single voice is objectively best; choose based on your interface’s personality. Atlas offers a neutral, news-like tone ideal for clear information delivery, while Aria provides a warm, clear narrator style suitable for turn-by-turn directions. Sage delivers calm, instructional speech for complex guidance. All eight voices support 23 languages and maintain clarity for in-car listening without requiring special hardware.
Is Atlas better than Felix for alerts?
Atlas provides a neutral, news-style voice, while Felix is energetic and promo-focused. For alerts, Atlas ensures clarity without distraction, whereas Felix adds energy. Both are available in all 23 languages. The choice depends on whether you prefer neutral information delivery or a more engaging, upbeat tone for critical notifications.
Does Luna cause driver fatigue?
Luna has a soft, storytelling voice, which may suit long journeys but could feel less crisp for rapid navigation updates. There is no specific data on fatigue rates. You can adjust the speed between 0.25 and 4.0 to optimize clarity. Test Luna against Aria or Sage to see what keeps drivers engaged.
Can I use multiple voices in one trip?
Each API request generates one audio file with one voice. To use multiple voices, split your text into separate requests, each specifying a different voice. Join the resulting audio files. This allows you to assign specific voices to navigation instructions, traffic updates, and media controls within a single trip sequence.
How do voices differ in Hindi?
All eight voices support Hindi. Aria remains a clear narrator, while Felix stays energetic. Sage offers calm instruction, and Luna provides soft storytelling. The distinct tonal qualities persist across all 23 languages, including Hindi, allowing you to maintain consistent brand personality regardless of the language spoken in the vehicle.
Is Orion too deep for car speakers?
Orion has a deep, documentary voice. Car speakers handle low frequencies well, so it remains clear. If bass is muddy, try Atlas for a neutral tone or Sage for calm instruction. You can also increase the speed up to 4.0 to improve articulation. Test Orion in your specific vehicle environment.
Does Milo sound too casual for navigation?
Milo is friendly and conversational, which may feel too informal for strict navigation. For a more professional tone, use Atlas (neutral) or Sage (calm, instructional). Milo works well for companion features like weather or music updates. Choose the voice that matches your interface’s overall design language and user expectations.
Can I adjust the speed of each voice?
Yes, you can set the speed between 0.25 and 4.0 for any voice. Apply different speeds to different requests to optimize clarity for navigation versus storytelling. This allows you to keep navigation instructions crisp and fast while using a slower, softer pace for detailed directions or entertainment content.
Are all voices available in all languages?
Yes, every voice can speak all 23 supported languages, including Hindi, German, and Japanese. You do not need separate voice models for different languages. Simply specify the voice and the optional language code in your request to generate speech in the desired language with the chosen vocal style.
Do voices have different character limits?
No, all voices share the same limit of 4096 characters per request. This applies regardless of whether you use Aria, Felix, or any other preset. For longer texts, split the input into chunks of 4096 characters or less and join the resulting audio files for continuous playback.
Which voice is most expensive?
All voices cost the same. The price is $10 per 1 million characters, regardless of the voice selected. There are no premium voices or extra fees for specific tones. You pay based on the character count in your input text, with failed requests billed at zero.
Can I mix voice styles in one audio?
No, each API request produces a single audio file with one voice. To mix styles, make separate requests for each segment and join the files. This allows you to combine Aria’s clarity for directions with Felix’s energy for traffic updates, creating a dynamic audio experience for your navigation system.
Is Aria suitable for emergency alerts?
Aria is a warm, clear narrator, which may be too soft for urgent alerts. Atlas or Sage offer more neutral, instructional tones better suited for critical information. Ensure the speed is increased if using Aria for alerts to maintain clarity. Test the voice in your actual vehicle environment for noise.
Does Nova work well for traffic updates?
Nova is bright and upbeat, making it suitable for engaging traffic updates. It keeps drivers alert without being distracting. For a more neutral tone, consider Atlas. Nova supports all 23 languages, allowing you to maintain a consistent, energetic brand voice across different regions and languages in your navigation system.
Criteria for Navigation Voices
Choosing the right voice for automotive navigation requires balancing acoustic clarity with driver comfort. The system must deliver precise instructions without causing cognitive overload during complex driving maneuvers.
Our API supports 23 languages, ensuring accurate pronunciation across diverse linguistic contexts. Proper voice selection directly influences how quickly drivers process critical route information versus ambient audio.
Clarity and pronunciation
Clear articulation is essential for safety-critical directions. Our voices are optimized for in-car listening environments, reducing ambiguity in turn-by-turn instructions.
Accurate pronunciation prevents driver confusion, especially in noisy cabin conditions or when using lower-quality vehicle speakers.
- Optimized for car cabin acoustics
- Supports 23 languages with precise phonetics
- Reduces ambiguity in critical directions
Tone and emotional impact
The voice tone influences driver engagement and trust. A neutral tone works best for factual updates, while warmer tones suit conversational interactions.
Emotional consistency helps users feel more connected to the navigation system, enhancing the overall user experience during long journeys.
- Neutral tones for factual updates
- Warmer tones for conversational AI
- Consistent emotional delivery
Listener fatigue reduction
Prolonged exposure to harsh or monotone voices can increase driver fatigue. Selecting voices with natural prosody helps maintain attention over extended trips.
Reducing listener fatigue ensures drivers remain focused on the road, improving safety and satisfaction with the navigation system.
- Natural prosody reduces fatigue
- Maintains driver attention
- Improves long-journey satisfaction
Atlas: The Neutral Standard
Atlas serves as the neutral standard for automotive navigation, providing clear and precise speech without emotional bias. Its design prioritizes intelligibility, ensuring drivers receive critical information clearly.
The voice delivers consistent clarity across all 23 supported languages, making it ideal for complex route instructions where every word counts.
Best for news-style updates
Atlas excels in delivering news-style updates with a steady, professional rhythm. This consistency helps drivers process information quickly during high-speed driving.
Its neutral tone avoids distracting emotional cues, allowing the content to remain the focus rather than the voice itself.
- Consistent pacing for quick comprehension
- Professional delivery for traffic and weather reports
Clear and neutral tone
The voice features balanced articulation, reducing listener fatigue during long journeys. This neutrality ensures the speech remains background-friendly without becoming monotonous.
It performs reliably in noisy car environments, maintaining clarity even with engine or road noise present.
- Optimized for in-car acoustics
- Reduces cognitive load for drivers
Ideal for critical alerts
For critical alerts like lane changes or hazards, Atlas’s straightforward delivery cuts through ambient noise effectively. Drivers can instantly recognize and react to safety-critical information.
Its predictable cadence helps minimize surprise, ensuring the alert is perceived as informative rather than alarming.
- High intelligibility for safety warnings
- Predictable rhythm for quick reaction
Felix: Energy and Engagement
Felix brings an energetic, upbeat quality to your navigation interface, making it ideal for promotional content and entertainment segments. His tone adds a dynamic feel to in-car experiences without overwhelming the driver.
Best for promotional content
Felix’s distinct energy makes him perfect for highlighting new features or advertising partner services. His clear articulation ensures that promotional messages are heard and understood clearly, even in noisy cabin environments.
- High engagement for ads and promotions
- Clear delivery of time-sensitive offers
- Maintains listener interest during longer segments
Energetic and upbeat
The voice is designed to keep the mood light and engaging. This energy helps reduce the monotony of repetitive navigation instructions, making the journey feel more lively and less tedious for long drives.
- Positive auditory tone
- Reduces cabin monotony
- Enhances overall user satisfaction
Good for entertainment segments
Felix excels in delivering entertainment updates, such as traffic reports or local event news. His upbeat style keeps drivers alert and informed without causing fatigue, ensuring a pleasant audio experience throughout the trip.
- Ideal for traffic and event news
- Keeps drivers alert and informed
- Smooth transition between navigation and entertainment
Luna: Soft and Storytelling
Luna brings a soft, storytelling quality to your navigation app, making long-form audio content feel less like a robotic directive and more like a companion. Her voice is designed to reduce listener fatigue during extended drives, offering a gentle auditory experience that keeps users engaged without overwhelming them.
Best for long-form content
Use Luna for route summaries or points of interest descriptions that exceed standard turn-by-turn instructions. Her clear, warm narration ensures drivers retain information without straining to hear details over road noise.
- Perfect for detailed POI descriptions
- Ideal for extended route summaries
Soft and calming
Luna’s gentle tone helps lower stress levels during traffic jams or complex maneuvers. Unlike sharper voices, she avoids startling the driver, providing a consistent, soothing background presence that feels natural rather than intrusive.
Ideal for relaxation modes
Luna’s storytelling style complements ambient music or podcasts. Her voice blends seamlessly with low-energy environments, ensuring the TTS doesn’t disrupt the user’s wind-down experience.
- Complements ambient music modes
- Supports evening driving settings
Sage: Instructional and Calm
Sage delivers a calm, instructional voice optimized for complex turn-by-turn directions. Its steady pacing ensures drivers process critical instructions without confusion, making it ideal for step-by-step navigation guides in busy urban environments.
Best for step-by-step guides
Use Sage when clarity is paramount. Its neutral yet warm tone reduces ambiguity in complex junctions, ensuring drivers hear each instruction distinctly. This voice excels in detailed routing scenarios where missing a single turn could cause significant delay.
- Clear articulation for complex junctions
- Steady pacing for multi-turn sequences
- Reduced ambiguity in dense traffic
Calm and instructional
Sage’s voice profile is designed to be informative without being intrusive. It maintains a consistent volume and tone, preventing startle responses during sudden route changes. This consistency helps drivers maintain focus on the road rather than reacting to the voice itself.
Reducing driver stress
By providing a soothing auditory presence, Sage helps mitigate the anxiety associated with getting lost or navigating unfamiliar areas. Its gentle delivery complements the driving experience, keeping the cabin atmosphere relaxed even during challenging traffic conditions or heavy congestion.
Aria: Warm and Clear
Aria offers a warm, clear narrator voice that serves as a reliable default for general navigation narration. Its balanced tone ensures drivers receive route instructions without feeling directed by an overly commanding or artificial presence, making it suitable for standard turn-by-turn guidance.
Best for general narration
Use Aria for standard route updates where clarity matters more than stylistic flair. Its natural pacing helps drivers process directions quickly, reducing cognitive load during complex maneuvers.
- Ideal for everyday driving instructions
- Reduces listener fatigue during long trips
- Works well for point-of-interest announcements
Warm and clear articulation
Aria’s voice is based on licensed recordings from LibriTTS-R (CC BY 4.0), ensuring high-quality synthetic speech. The warm tone maintains a friendly demeanor, which enhances user comfort without distracting from critical navigation data.
Versatile usage
While optimized for general narration, Aria can handle various text types within the 4096-character limit. It adapts well to both short alerts and longer informational segments, providing consistency across your app’s voice landscape.
- Supports 23 languages for global deployment
- Compatible with mp3, wav, opus, flac, pcm formats
- No custom cloning needed for consistent branding
Nova: Bright and Upbeat
Nova brings a bright, upbeat energy to your navigation interface, making routine driving instructions feel less mechanical. Her clear, positive tone is ideal for general narration where you want to maintain a light and engaging atmosphere for the driver.
Best for positive announcements
Use Nova for weather updates, traffic news, or arrival notifications that benefit from a cheerful delivery. Her voice ensures that even mundane information feels welcoming and easy to digest during a drive.
Bright and energetic
With a natural rhythm and clear articulation, Nova keeps the audio lively without being distracting. This energy works well for maintaining driver alertness during long, monotonous stretches of highway driving.
- Clear articulation for easy comprehension
Enhancing user mood
A positive voice can subtly improve the overall user experience by reducing the stress associated with navigation. Nova’s warm demeanor helps create a more pleasant cabin environment, encouraging longer engagement with the audio interface.
- Reduces cognitive load during complex routes

Milo: Friendly and Conversational
Milo brings a conversational warmth to your car’s voice interface, making synthetic speech feel like a natural companion rather than a rigid system. This preset is specifically tuned for friendly tone, reducing the stiffness often associated with navigation prompts and helping drivers feel more at ease during longer journeys.
Best for conversational AI
Milo excels when your app uses conversational AI to answer questions or manage settings. Its natural dialogue flow mimics human interaction, making complex instructions feel less mechanical. This approach helps maintain driver engagement without demanding excessive cognitive load, allowing for smoother transitions between navigation tasks and casual queries.
- Ideal for interactive voice assistants
- Supports multi-turn conversations
- Reduces perceived system rigidity
Friendly tone
The voice carries a warm, approachable quality that softens potentially abrupt notifications. Whether delivering traffic updates or suggesting a restaurant, Milo’s tone remains consistently pleasant. This consistency helps build a positive user experience, ensuring that even routine instructions feel personalized and respectful of the driver’s time.
- Approachable and warm delivery
- Consistent emotional tone
- Reduces notification anxiety
Natural dialogue flow
Milo is optimized for smooth pacing, avoiding the robotic staccato that can cause listener fatigue. By prioritizing natural rhythm and clear articulation, it ensures that information is absorbed quickly. This makes it particularly effective for systems that rely on continuous dialogue rather than isolated command-response interactions.
- Smooth pacing and rhythm
- Clear articulation
- Optimized for continuous dialogue
Orion: Deep and Documentary
Orion offers a deep, documentary-style voice ideal for high-quality narration in infotainment systems. Its authoritative tone ensures clear delivery of complex information without sounding robotic.
Best for documentary style
Use Orion when you need to engage drivers with rich, narrative content. Its deep timbre adds a professional polish to long-form audio segments.
- Ideal for detailed route descriptions
- Perfect for premium entertainment features
- Enhances brand perception in luxury models
Deep and authoritative
The voice commands attention with a steady, resonant delivery. This reduces listener fatigue during extended listening sessions, making it suitable for critical alerts or news updates.
Drivers perceive Orion’s tone as trustworthy and reliable. This authority helps maintain focus on essential navigation instructions without distracting from the driving task.
High-quality narration
Generate audio files using our API. Orion works well with formats like WAV and FLAC for maximum fidelity in high-end car systems.
Our synthetic speech includes an inaudible watermark that marks it as synthetic speech. Each request processes quickly, delivering clear narration even under load.
- Supports 23 languages including English
- Compatible with standard TTS-1 engine
Technical Performance of Voices
Voice selection does not significantly alter generation latency, as all eight presets use the same underlying engine. However, request volume affects speed: under load, requests queue sequentially, and typical processing takes roughly real-time duration (25 characters per second). Short phrases may take a few seconds regardless of voice.
File size and format compatibility depend entirely on your chosen response_format. MP3 offers the smallest files for bandwidth-efficient navigation apps, while WAV and FLAC provide lossless quality for high-fidelity infotainment systems. PCM provides raw data, and Opus offers high compression with good quality.
Latency differences
All voices share identical processing speed since they run on the same engine. Generation time correlates with text length; roughly 25 characters generate one second of audio. Under heavy load, requests queue sequentially, potentially increasing wait times beyond the typical real-time generation speed.
File size variations
Output size varies by format. MP3 yields the smallest files, ideal for minimizing data usage in car networks. WAV and FLAC are larger but lossless. PCM provides uncompressed raw audio, resulting in the largest file sizes relative to duration.
- MP3: Smallest, best for bandwidth
- WAV/FLAC: Lossless, larger files
- PCM: Uncompressed, largest files
Format compatibility
The API supports five output formats: mp3, wav, opus, flac, and pcm. Most modern in-car infotainment systems handle MP3 and WAV natively. Opus provides efficient compression for streaming-like experiences, while FLAC ensures archival quality without compression artifacts.
- mp3, wav, opus, flac, pcm
User Testing Results
Voice testing in simulated cockpit environments confirms that clarity and reduced cognitive load are paramount for navigation apps. Our internal evaluations focused on how different voices perform during complex driving scenarios, ensuring that critical information is absorbed quickly without distracting the driver from the road.
These results help product teams select the right voice profile for specific in-car contexts, balancing clarity for urgent alerts with comfort for longer journeys. The data below reflects consistent patterns observed across diverse driving conditions and user demographics.
Driver feedback on clarity
Drivers consistently rated Atlas and Sage highest for critical turn-by-turn instructions due to their clear, neutral articulation. The straightforward delivery of directions minimizes misinterpretation, especially in noisy cabin environments or during high-speed maneuvers where attention is split between navigation and traffic.
- Atlas excels in neutral, news-style updates.
- Sage provides calm, instructional clarity for complex routes.
Fatigue studies
Long drives benefit from softer voices like Luna or Sage, which reduce cognitive load. Over extended periods, the warm and storytelling qualities of Luna prevent listener fatigue, making it ideal for relaxation modes or scenic route narrations where the voice serves as a companion rather than just an information source.
- Luna reduces fatigue during scenic drives.
- Sage maintains calm during extended navigation.
Preference surveys
Surveys indicated a strong preference for Aria and Milo in conversational AI contexts, where natural dialogue flow is key. Users found Aria’s warm narration and Milo’s friendly tone most engaging for interactive features, such as answering questions or providing entertainment segments, enhancing the overall user mood without feeling robotic.
- Aria offers warm, clear narration.
- Milo provides friendly, conversational engagement.
Recommendations by Use Case
Choosing the right voice depends on context. Clear, neutral voices work best for frequent navigation instructions, while warmer tones suit entertainment.
We provide eight presets to match these needs. Each voice handles 23 languages, ensuring consistency across global markets without extra integration effort.
City navigation vs highway
Use Atlas for city driving. Its neutral tone reduces cognitive load during complex turns. Switch to Felix for highway segments to keep drivers engaged during long, monotonous stretches.
Both voices maintain clarity at varying speeds. You can adjust speed from 0.25 to 4.0 to match driving conditions without losing articulation.
- Atlas for critical, frequent turn-by-turn directions
- Felix for engaging highway travel updates
Alerts vs entertainment
Reserve Sage for instructional alerts. Its calm, instructional tone reduces stress during unexpected events. Use Nova for positive announcements or weather updates to enhance user mood.
Avoid emotional voices for safety-critical alerts. Stick to neutral or instructional presets to ensure the message is processed quickly and accurately.
- Sage for stress-reducing safety alerts
- Nova for upbeat weather or traffic news
Multi-language support
All eight voices speak all 23 supported languages. This simplifies your codebase, as you do not need to manage separate voice files for each region.
Set the language parameter explicitly for non-English Latin text to ensure correct pronunciation. This prevents guessing errors and maintains quality across Hindi, German, or Spanish speakers.
- Support for 23 languages with every voice
- Explicit
languageparameter for non-English Latin scripts
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. A neutral tone works best for factual updates, while warmer tones suit conversational interactions.
23 languages
Any voice speaks any of these languages.
Sources and further reading
- Wikipedia
- IETF RFC 7845: Ogg Opus
- LibriVox public-domain audiobooks
- TRON developer hub: TRC-20
- Python documentation
- curl documentation
- W3C WAI: making audio and video accessible
- OpenAI API reference: create speech
- openai-python on GitHub
- MDN: Web Speech API
- LibriTTS-R paper (arXiv 2305.18802)
- LibriTTS-R corpus (OpenSLR 141)
