Frequently asked questions
Which voice is best for transit apps?
Atlas offers a neutral, news-style tone suitable for clear station announcements. For a warmer, conversational feel suitable for app notifications, try Milo or Luna. Each voice has a distinct character, so test them with your specific content.
Can I mix voices in one announcement?
The API generates one audio file per request, so you cannot mix voices in a single file directly. To achieve this, generate separate audio clips for each voice segment and join them in your app or using a simple audio tool. This keeps each request under 4096 characters.
Do voices sound different in languages?
All eight voices support 23 languages, including English, Spanish, and Japanese. While the voice timbre remains consistent, the pronunciation and intonation adapt to the language. If you do not specify the language, the system guesses it, but setting it explicitly ensures better accuracy for non-English Latin text.
Is MP3 better than WAV for apps?
MP3 is often preferred for transit apps due to its smaller file size and wide compatibility. WAV offers lossless quality but results in larger files, which may slow down loading times on mobile networks. Choose MP3 for efficient delivery and WAV only if high fidelity is critical for your use case.
How do I choose a voice tone?
Select voices based on your content’s purpose. Use Felix for energetic promos or Sage for calm instructions. For standard announcements, Atlas provides a neutral tone. Test each voice with your actual script to ensure the tone matches your brand’s personality and remains clear in various listening environments.
Can I change the voice after generation?
No, the voice is fixed once the audio file is generated. To change the voice, you must make a new API request with the desired voice parameter. This process is fast, and failed requests are free, so you can experiment with different voices without incurring extra costs.
Do voices sound natural in noisy stations?
The voices are designed for clarity, but background noise can affect perception. Use a neutral voice like Atlas for maximum intelligibility. Ensure your playback volume is appropriate for the environment. The API does not adjust audio dynamically, so clarity depends on your app’s audio output settings and the environment.
Is there a standard voice for announcements?
There is no single official standard, but Atlas is optimized for a neutral, news-style delivery common in transit systems. Many teams find Felix suitable for promotional announcements due to its energetic tone. You can choose any of the eight voices based on your specific branding and content needs.
How do I test voice clarity?
Generate short audio clips using the API and play them back in your target environment. Check for clear pronunciation and appropriate speed (adjustable from 0.25 to 4.0). Use the free trial credit to test multiple voices and languages before scaling up your integration with the full API.
Can I use multiple voices in one file?
The API returns one audio file per request, so you cannot embed multiple voices in a single file directly. You must generate separate files for each voice and concatenate them in your application. This approach gives you full control over timing and transitions between different voice segments.
Do voices support special characters?
The API accepts standard text input up to 4096 characters. Special characters like punctuation and emojis are generally handled, but complex symbols may affect pronunciation. Test your specific text to ensure the output matches your expectations. Spaces and punctuation count toward the character limit and pricing.
How do I handle language switches?
Specify the ISO language code in the request body to ensure accurate pronunciation. If omitted, the system guesses the language, which may default to English for Latin text. Explicitly setting the language prevents errors when switching between supported languages like English, Japanese, or Spanish within the same session.
Is the API good for real-time updates?
The API is suitable for near-real-time updates, but generation takes a few seconds for short phrases. Under load, requests may queue, so plan for slight delays. For critical real-time needs, consider pre-generating common announcements. The API processes requests sequentially, so high concurrency may increase latency.
Understanding Voice Tones
Choosing the right voice tone sets the user experience for your transit app. Warm narrators like Aria provide a clear, steady presence for general route information. Bright upbeats, such as Nova, inject energy into app notifications.
Deep documentaries, represented by Orion, offer a commanding authority suitable for safety warnings or major service changes. The distinct tonal differences help users instantly recognize the urgency or context of an announcement without needing to read the text carefully.
Warm narrators
Aria delivers a soft, storytelling quality that works well for detailed route explanations or station history segments. Her clear enunciation ensures that complex transit instructions remain understandable, even in noisy environments.
- Aria: warm, clear narrator
- Best for: detailed route info
- Style: storytelling, clear
Bright upbeats
Nova brings a bright, upbeat energy ideal for quick app notifications or promotional announcements. Her tone feels modern and active, making routine updates feel less like interruptions and more like helpful tips.
- Nova: bright, upbeat
- Best for: app notifications
- Style: energetic, modern
Deep documentaries
Orion provides a deep, documentary-style voice that conveys seriousness and importance. Use this tone for safety alerts, emergency delays, or critical service changes where you need to command attention immediately.
- Orion: deep, documentary
- Best for: safety alerts
- Style: serious, commanding
Matching Voice to Context
Matching voice tone to transit context ensures passengers understand critical information without cognitive overload. A mismatched tone can cause confusion or anxiety, especially during high-stress moments like sudden delays or platform changes.
Station delays
Use Aria or Luna for delay announcements. Their warm, clear tones soften the blow of unexpected waits, making routine disruptions feel manageable rather than alarming.
- Avoid Felix’s high energy; it may feel too promotional for bad news.
- Keep text under 4096 characters to ensure immediate playback.
Safety warnings
Atlas provides a neutral, authoritative news-style delivery ideal for safety alerts. His consistent pacing ensures clarity in noisy station environments without sounding robotic.
- Ensure language is explicitly set for non-English warnings to avoid guessing errors.
- Use WAV format for critical alerts to preserve audio fidelity in loud areas.
Route information
Milo’s conversational tone works well for standard route updates, making navigation feel like a friendly guide. For complex transfers, Sage’s instructional calm helps passengers follow multi-step directions.
- Split long route descriptions into separate requests if they exceed character limits.
- Test audio clarity at 1.0x speed before increasing for faster throughput.
Language Considerations
ClearCast’s text-to-speech engine supports 23 languages, allowing transit apps to serve diverse passenger bases without needing separate voice actors. The system automatically detects language from the script, but explicit language codes are recommended for non-English Latin text to ensure accurate pronunciation.
Consistent language handling prevents awkward accents in announcements. For example, setting the language parameter explicitly ensures Spanish warmth in Madrid or Arabic clarity in Dubai, rather than defaulting to English pronunciation rules.
English clarity
English is the default for Latin-script text if no language is specified. Voices like Aria and Felix deliver crisp, natural-sounding English for station delays and safety warnings. This default behavior simplifies integration for single-language markets.
- Default language for Latin text if omitted
- High clarity for standard announcements
- Supports all 8 voice presets
Spanish warmth
Spanish is fully supported with natural intonation across all voices. Use Nova for upbeat route updates or Sage for calm instructional guidance. The engine handles complex Spanish phonemes accurately, ensuring passengers understand critical information.
- Full support for all 8 voices
- Natural intonation for route info
- No extra configuration needed
Supported languages
We support 23 languages including German, French, Japanese, Korean, Hindi, and Portuguese. Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, and Bengali are not supported. If your route serves these areas, consider pre-generating audio or using a different provider for those specific stops.
- 23 languages supported
- Not supported: Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, Bengali
- Set language parameter for non-English Latin text
Technical Format Choices
Choosing the right audio format balances file size, quality, and playback compatibility. Our API supports MP3, WAV, Opus, FLAC, and PCM, letting you tailor output to your specific infrastructure needs without compromising clarity.
For transit apps, efficient delivery often matters more than studio-grade fidelity. Selecting the optimal format ensures announcements load quickly on mobile devices and play smoothly across varied digital signage hardware.
MP3 compression
MP3 offers a balance between file size and acceptable quality for most transit use cases. It reduces bandwidth usage significantly while maintaining clear speech intelligibility, making it ideal for on-demand API calls where storage and transfer costs matter.
- Supports variable bitrates for optimal efficiency
- Widely compatible with all mobile and web players
- Ideal for dynamic route announcements
WAV quality
WAV provides uncompressed, high-fidelity audio perfect for local playback or high-quality digital signage. While files are larger, the absence of compression artifacts ensures crisp sound in environments where audio clarity is critical, such as quiet station platforms.
- Lossless quality for maximum fidelity
- No decoding overhead for immediate playback
- Best for high-end digital displays
Opus efficiency
Opus delivers audio quality at low bitrates, making it suitable for bandwidth-constrained mobile networks. Its efficiency allows faster delivery of announcements to users on slow connections, ensuring timely updates without excessive data consumption.
- Superior quality at low bitrates
- Low latency for real-time updates
- Optimized for mobile network conditions
Speed and Timing
Speed controls how quickly text is spoken, adjustable from 0.25 to 4.0. This setting directly impacts clarity and user comprehension in noisy transit environments.
Choosing the right speed ensures announcements are heard clearly without sounding rushed or dragging. It balances listener comfort with operational efficiency.
Faster announcements
Higher speeds (e.g., 2.0–4.0) are useful for brief, non-critical updates like route changes or minor delays. They save time but may reduce comprehension in loud stations.
- Best for non-urgent informational updates
- Reduces total generation time
- May sacrifice clarity in noisy areas
Slower clarity
Lower speeds (0.25–0.75) enhance understanding for complex instructions or safety warnings. Ideal for elderly passengers or those with hearing impairments.
Use slower speeds for critical information where every word matters, such as emergency alerts or platform changes.
- Improves comprehension for complex messages
- Ideal for safety and emergency alerts
- Better for noisy or crowded environments
Balanced pace
A default speed of 1.0 offers a natural, conversational tone suitable for most routine announcements. It strikes a balance between clarity and efficiency.
- Default setting for general use
- Natural, conversational tone
- Suitable for most route information
Voice Preset Overview
Selecting the right voice tone ensures your transit announcements resonate with passengers without causing fatigue. Our engine provides distinct presets designed for specific auditory environments, from busy platforms to quiet app interfaces.
Aria for stories
Aria offers a warm, clear narration style ideal for detailed route explanations or safety stories. Her tone is engaging yet professional, making complex information easier to digest during longer delays.
Nova for energy
Nova brings a bright, upbeat quality perfect for peak-hour updates or promotional messages. Her energetic delivery keeps listeners alert and helps convey urgency without sounding aggressive or stressful.
Luna for softness
Luna delivers a soft, storytelling voice suited for gentle reminders or evening service updates. Her calming presence reduces auditory stress in noisy stations or quiet app notifications.
Male Voice Options
ClearCast offers three distinct male voices designed to handle the varied auditory landscape of modern transit systems. Each preset delivers a specific tone, ensuring your announcements fit the context perfectly, whether you are broadcasting urgent route updates or playing friendly in-app notifications.
Felix for promos
Felix brings an energetic and upbeat quality that cuts through ambient station noise. His tone is ideal for promotional announcements or short, attention-grabbing alerts that need to feel lively without sounding aggressive.
- High energy for marketing messages
- Clear articulation in busy environments
- Upbeat pacing for short clips
Milo for chats
Milo provides a friendly, conversational voice that feels natural for app notifications or interactive guides. His approachable tone helps users feel guided rather than commanded, making routine updates feel personal and clear.
- Conversational and approachable
- Ideal for app notifications
- Natural pacing for dialogue
Atlas for news
Atlas delivers a neutral, authoritative tone perfect for real-time station delays or safety warnings. His consistent, professional delivery ensures critical information is heard as factual and reliable, reducing ambiguity during peak travel hours.
- Neutral and authoritative
- Clear for delay announcements
- Professional safety warnings

Female Voice Options
ClearCast offers distinct female voices tailored for transit communication, ensuring your announcements resonate with passengers across different contexts. Whether you need calm instructions or energetic updates, selecting the right preset enhances user experience without overwhelming the audio environment.
Each voice carries a specific tone derived from licensed recordings, providing a consistent brand identity for your station announcements and app notifications.
Sage for instruction
Sage delivers a calm, instructional tone ideal for step-by-step guidance. Her clear enunciation works well for complex route changes or boarding instructions, ensuring passengers understand critical information without confusion.
Use Sage when conveying detailed operational updates where clarity outweighs emotional expression. Her steady pace helps reduce cognitive load for travelers navigating unfamiliar systems.
- Best for: detailed boarding instructions
- Ideal for: complex route changes
- Tone: calm and instructional
Orion for depth
Orion provides a deep, documentary-style voice that adds authority to safety warnings or major service alerts. While often categorized as male, Orion's rich depth offers a distinct auditory marker for high-priority announcements.
- Best for: safety warnings
- Ideal for: major service alerts
- Tone: deep and authoritative
Nova for brightness
Nova brings a bright, upbeat energy perfect for promotional messages or positive updates like new line openings. Her clear, energetic delivery keeps passengers engaged and informed about exciting developments.
- Best for: promotional messages
- Ideal for: new line announcements
- Tone: bright and upbeat
Testing Your Selection
Validating voice choices ensures your transit announcements cut through background noise while remaining clear to passengers. Since requests are processed sequentially, testing helps you balance clarity with the generation speed constraints of the API.
A/B testing voices
Generate short clips using different voices like Aria for route info and Felix for promos. Compare their intelligibility in simulated environments.
- Test 4096-character limits with typical station delay scripts.
- Compare Aria’s warmth against Nova’s brightness for alerts.
User feedback loops
Collect feedback from passengers in quiet vs. noisy areas. Use this data to refine which voices work best for specific announcements.
Noise environment tests
Play audio through your digital signage speakers at varying volumes. Ensure the tone remains crisp even with high ambient noise levels.
- Test Opus files for efficiency in noisy conditions.
- Verify clarity of slower speeds for elderly passengers.
Integration Tips
Efficient audio delivery relies on smart caching strategies. Pre-generating common announcements reduces latency during peak hours. Storing popular route updates locally ensures instant playback even when network conditions are poor.
Dynamic updates require careful handling of character limits. Split long texts into chunks under 4096 characters. Join the resulting audio files seamlessly for continuous playback.
Pre-generate content
Generate audio files for static announcements like station names or standard safety warnings. Store them in your CDN. This reduces API calls and ensures consistent playback speed during rush hour when request queues might be longer.
- Store static announcements in a CDN
- Pre-generate for high-frequency routes
Cache popular routes
Cache audio for frequently requested routes. This minimizes API usage and ensures immediate playback. Since requests are processed sequentially, caching helps manage queue delays during high traffic periods.
Dynamic updates
For real-time delays, send new text to the API. Split long messages into parts under 4096 characters. Join the resulting audio files for a continuous announcement. Monitor character counts to avoid truncation.
Common Mistakes to Avoid
Transit environments are rarely quiet. Rush hour stations and moving vehicles introduce significant background noise that can mask subtle audio cues. If your announcements are too soft or lack clarity, passengers will miss critical updates regardless of voice quality.
Equally important is how you structure your text. The API limits input to 4096 characters per request, but even within that limit, overly dense sentences can become difficult to understand when spoken aloud. Concise phrasing ensures better comprehension.
Ignoring noise levels
Ambient noise in transit hubs often exceeds 70 dB. Ensure your audio files have sufficient gain and use voices like Atlas or Sage that cut through background chatter effectively. Avoid overly soft tones like Luna for urgent alerts in noisy areas.
- Test audio at high volume to simulate station environments
- Use neutral or crisp voices for urgent alerts
- Avoid soft storytelling tones for critical safety warnings
Overloading text
Long, complex sentences can confuse listeners. Break down route changes or safety instructions into short, direct statements. This not only improves clarity but also stays within the 4096-character limit without requiring complex splitting logic.
- Keep sentences under 20 words for clarity
- Use active voice for direct instructions
- Limit special characters to avoid processing errors
Wrong language setting
The API guesses language from script, which fails for Latin-script languages other than English. If you broadcast in Spanish or French, explicitly set the language field. Otherwise, the system may default to English, resulting in unnatural pronunciation.
- Explicitly set the
languageparameter for non-English text - Verify supported languages: 23 total, excluding Vietnamese and Thai
- Test edge cases like mixed-language announcements
Future-Proofing Your Audio
Future-proofing your audio strategy ensures your transit app remains reliable as ridership grows and technology evolves. By building a scalable architecture, you can handle increased request volumes without degrading the quality of station announcements or safety alerts.
Consistent branding keeps users engaged by maintaining a recognizable voice across all platforms, from mobile notifications to digital signage. This consistency builds trust and ensures that critical information is always delivered in a tone that matches your system’s identity.
Scalable architecture
Design your integration to handle spikes in demand, such as rush hour or unexpected delays. Use the API’s straightforward JSON structure to generate audio on the fly or pre-generate popular routes.
Ensure your system can process multiple requests concurrently while staying within the 60 requests per minute limit per key.
- Pre-generate audio for static route info
- Cache dynamic delay announcements
- Monitor usage to avoid rate limits
Consistent branding
Select a single voice preset, such as Aria or Atlas, for all general announcements. This creates a familiar auditory experience for your riders, reinforcing your brand’s reliability.
Ensure that safety warnings and route updates maintain the same tone to avoid confusing users with inconsistent voice personalities.
Easy updates
Store your generated audio files in a scalable storage solution. Update announcements quickly when routes change or new stations open.
Use the API’s speed control to adjust pacing for different contexts, ensuring clarity without needing new voice recordings.
- Update audio via API
- Adjust speed for clarity
- Maintain voice consistency
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Aria delivers a soft, storytelling quality that works well for detailed route explanations or station history segments.
23 languages
Any voice speaks any of these languages.
