Frequently asked questions
Which voice is best for formal transit announcements?
Atlas is the neutral, news-style voice ideal for formal transit announcements. It provides a clear, professional tone suitable for station updates. Alternatively, Sage offers a calm, instructional style for detailed route information. Both voices support all 23 languages, ensuring consistent quality across your network without requiring multiple voice profiles.
Can I use the same voice for all languages?
Yes, all eight voices can speak any of the 23 supported languages. You can use a single voice preset, such as Atlas or Aria, across English, Spanish, and Arabic. If the language is not set, the system guesses it from the script, but explicitly setting the language code ensures accurate pronunciation.
Is there a difference between tts-1 and tts-1-hd voices?
The tts-1 and tts-1-hd models use the same engine and voices, differing only in audio quality. Both incur the same price of $10 per million characters. Choose hd for clearer audio in quiet stations or tts-1 for standard efficiency. The voice characteristics remain identical between models.
Do voices sound different in Arabic compared to English?
The voice timbre remains consistent, but pronunciation adapts to the language. Each voice is trained to handle phonetic nuances across all 23 supported languages. For example, Aria sounds natural in both English and Arabic. Ensure you specify the language parameter to avoid incorrect guessing, especially for Latin-script languages like Turkish or Malay.
Can I combine multiple voices in one announcement?
No, each API request generates audio for a single voice. To combine voices, you must make separate requests for each segment and join the resulting audio files. The API returns individual audio files that you can concatenate using standard audio processing tools before broadcasting to your transport network.
Is the Atlas voice suitable for emergency announcements?
Atlas provides a neutral, news-style tone that conveys authority and clarity, making it suitable for emergency announcements. Its calm delivery ensures passengers understand critical information without distraction. You can adjust the speed up to 4.0 if faster delivery is needed during urgent situations, while maintaining intelligibility across all supported languages.
Which voice is best for long route descriptions?
Luna is designed for soft, storytelling narration, making it ideal for long route descriptions. Sage is also suitable for instructional content. Both voices maintain clarity over longer texts. Since texts are split into 4096-character parts, these voices ensure consistent quality throughout extended announcements without sounding fatigued or monotonous for passengers.
Can I adjust the tone of a voice after generation?
No, you cannot adjust the tone after generation. The API returns a static audio file. You can adjust the speed between 0.25 and 4.0 during the request. For different tonal qualities, select a different voice preset like Aria for warmth or Felix for energy. Post-processing is required for further changes.
Do voices have accents or regional variations?
The voices do not have specific regional accents. If you need specific regional variations, you must select the appropriate voice preset that best matches your desired tone. The system supports 23 languages with consistent voice characteristics.
Is the Luna voice too slow for busy stations?
Luna is described as soft and storytelling, but its speed is adjustable. You can increase the speed up to 4.0 to suit busy station environments. The default speed is suitable for calm settings, but you can configure it to match the pace of your announcements without changing the voice preset.
Can I use the API for live bus announcements?
Yes, but with latency considerations. Requests are processed sequentially, so under load, a request may wait in a queue. For live announcements, ensure your system handles the delay. Short phrases take a few seconds to generate. You can pre-generate common announcements to ensure immediate playback during live operations.
Is the Nova voice too energetic for formal settings?
Nova is bright and upbeat, which may be too energetic for formal settings. For formal announcements, Atlas or Sage are better choices. Atlas offers a neutral news style, while Sage provides a calm instructional tone. These voices maintain a professional demeanor suitable for corporate transport environments without the high energy of Nova.
Do voices support SSML for emphasis?
No, the API does not support SSML. You cannot add emphasis or pauses using SSML tags. To control pacing, adjust the speed parameter between 0.25 and 4.0. For specific emphasis, you may need to split the text into separate requests or use punctuation, though punctuation does not guarantee pauses.
Can I clone a specific announcer's voice?
No, custom voice cloning is not offered. The API uses eight fixed presets based on licensed recordings. You cannot upload your own voice or create a custom clone. If you need a specific sound, select the preset that best matches your requirements, such as Felix for energetic or Orion for deep tones.
Is the API suitable for radio broadcasting?
Yes, the API generates audio files suitable for radio broadcasting. You can use the tts-1-hd model for higher fidelity audio. The output formats include mp3, wav, opus, flac, and pcm. Each file contains an inaudible watermark marking it as synthetic speech, which may be relevant for broadcast compliance.
Can I use the API for mobile apps?
Yes, you can integrate the API into mobile apps using standard HTTP requests or the OpenAI SDK. Change the base_url to https://api.paisita.info/v1. The API returns audio files that can be played directly in your app. Ensure your app handles network latency and queues requests appropriately for a smooth user experience.
Do voices support background music?
No, the API generates voice audio only. It does not support mixing background music. You must add background music using post-processing tools or your application's audio engine after receiving the audio file from the API. The API focuses solely on converting text to speech with the selected voice and speed.
Is the API secure for passenger data?
Input text is used only to produce audio and is not used for training. Generated audio is returned in the response and not stored. Request metadata is kept for billing. The API uses standard Bearer token authentication. Ensure you protect your API key and transmit data over HTTPS to maintain security.
Can I use the API for multilingual announcements?
Yes, all voices support 23 languages. You can switch languages by setting the language parameter in the request. The system supports Arabic, Chinese, English, Spanish, and others. Explicitly setting the language ensures correct pronunciation, especially for Latin-script languages where the system might otherwise guess English. This simplifies deployment across mixed-language regions.
Is the API free for small transport fleets?
No, there is no free tier. New accounts receive a $0.50 trial credit valid for 7 days. After that, you pay $10 per million characters. You can top up using USDT on the TRON network. There are no monthly plans or minimum spends, so you only pay for what you use.
How do I troubleshoot errors in voice selection?
Ensure the voice name is one of the eight presets: aria, nova, luna, sage, felix, milo, atlas, or orion. Check that the language parameter matches the text if specified. Verify your API key is active. If errors persist, check the request format and ensure the text is under 4096 characters.
Can I use the API for emergency announcements?
Yes, the API is suitable for emergency announcements. Use Atlas for a neutral, authoritative tone or Felix for energetic delivery. You can increase the speed up to 4.0 for faster alerts. Ensure your system handles queue delays during high load. The API provides reliable audio generation for critical transport updates.
Is the API accessible from mobile devices?
Yes, the API is accessible via standard HTTP requests from any device with internet connectivity. You can use it from mobile devices by integrating the OpenAI SDK or making direct POST requests. Ensure your mobile app handles network latency and queues requests appropriately to maintain a smooth user experience for passengers.
Can I use the API for radio broadcasting?
Yes, the API generates audio files suitable for radio broadcasting. You can use the tts-1-hd model for higher fidelity audio. The output formats include mp3, wav, opus, flac, and pcm. Each file contains an inaudible watermark marking it as synthetic speech, which may be relevant for broadcast compliance.
Do I need to pay extra for Arabic language support?
No, there is no extra charge for Arabic language support. All 23 languages are included in the standard pricing of $10 per million characters. You can use any voice preset with any supported language without additional fees. The only cost is based on the character count of your text input.
How do I renew my API key?
You can issue a new API key at any time from your account dashboard. When a new key is issued, the old one stops immediately. This allows you to rotate keys for security without downtime. You do not need to update your integration settings if you continue using the same key.
Can I use the API for emergency announcements?
Yes, the API is suitable for emergency announcements. Use Atlas for a neutral, authoritative tone or Felix for energetic delivery. You can increase the speed up to 4.0 for faster alerts. Ensure your system handles queue delays during high load. The API provides reliable audio generation for critical transport updates.
Is the API secure for passenger data?
Input text is used only to produce audio and is not used for training. Generated audio is returned in the response and not stored. Request metadata is kept for billing. The API uses standard Bearer token authentication. Ensure you protect your API key and transmit data over HTTPS to maintain security.
Can I integrate the API with my existing transport software?
Yes, the API uses a standard JSON request shape compatible with the OpenAI SDK. You can integrate it by changing the base_url to https://api.paisita.info/v1. It works with any software that can send HTTP POST requests. This simplifies deployment across your existing transport management systems.
Do you offer technical support for API integration?
Support is available by email only. There is no phone number or live chat. You can contact us for help with integration issues, voice selection, or billing. Response times may vary. Ensure you provide your API key and request details when contacting support to expedite assistance with your transport system integration.
Voice profiles overview
Paisita Voices provides eight distinct voice presets designed for consistent transport announcements across mixed-language fleets. Each voice is built from licensed recordings, ensuring a reliable audio standard for public address systems. The selection covers a range of tones, from energetic promotions to calm instructional updates, allowing teams to match the voice to the context of the announcement.
Female voice options
The female presets include Aria for clear narration, Nova for upbeat energy, Luna for soft storytelling, and Sage for calm instruction. These voices are suitable for everything from route details to safety protocols.
- Aria: Warm, clear narrator for general updates.
- Nova: Bright, upbeat tone for promotions.
- Luna: Soft, storytelling style for long routes.
- Sage: Calm, instructional delivery for safety.
Male voice options
Male voices offer Felix for energetic promos, Milo for conversational service updates, Atlas for neutral news, and Orion for deep, documentary-style information. These options provide distinct auditory cues for different types of information.
- Felix: Energetic, ideal for short blasts.
- Milo: Friendly, conversational tone.
- Atlas: Neutral, news-style clarity.
- Orion: Deep, authoritative delivery.
Neutral tone selection
For fleets requiring a consistent, non-distracting background voice, Atlas provides a neutral standard. This preset is designed for news-style announcements where clarity and professionalism are prioritized over emotional tone, making it ideal for mixed-language regions.
Atlas: The neutral standard
Atlas provides a neutral, news-style voice ideal for consistent fleet announcements. Its clear and professional tone ensures messages are delivered without emotional bias, making it suitable for routine transport updates across large teams.
Ideal for news-style announcements
Atlas delivers straight-forward information with a steady rhythm, mimicking standard broadcast delivery. This consistency reduces listener fatigue during repeated loop announcements, ensuring passengers receive critical route or schedule changes clearly.
- Consistent delivery for repeated schedule updates
- Neutral tone reduces listener fatigue
- Ideal for time-sensitive operational changes
Clear and professional tone
The voice emphasizes articulation over expressiveness, ensuring high clarity even in noisy environments. It avoids dramatic inflections, focusing instead on precise enunciation of destination names and times.
- High articulation for noisy stations
- Minimal emotional inflection
- Precise enunciation of names and times
Best for mixed-language fleets
Atlas speaks all 23 supported languages, making it the safest choice for diverse routes. When the language parameter is unspecified, it defaults to English, but can be explicitly set for Arabic, Chinese, or other languages without changing the voice profile.
- Supports all 23 languages
- Consistent quality across languages
- Easy language switching via API
Milo: The friendly option
Milo delivers a friendly, conversational tone ideal for customer service updates. His voice sounds like a helpful assistant rather than a rigid system, making routine announcements feel personal and less mechanical.
Conversational tone
Milo’s natural phrasing reduces listener fatigue during repeated messages. This approachability helps maintain passenger attention without sounding overly formal or distant.
Good for customer service updates
Use Milo for real-time service changes, such as gate changes or minor delays. His clear yet warm delivery ensures passengers perceive the information as helpful rather than authoritative.
- Gate change notifications
- Minor delay explanations
- Welcome messages at stations
Approachable for passengers
Passengers respond better to Milo’s tone during high-traffic periods. His style reduces perceived stress when communicating important but routine logistical details.
Luna: The soft storyteller
Luna offers a soft, storytelling voice ideal for conveying detailed route information. Her calming tone reduces passenger anxiety during unexpected delays, providing clear context without sounding mechanical.
Ideal for detailed route info
Luna’s narrative style excels when explaining complex stop sequences or detours. Her clear articulation ensures passengers understand specific geographic changes without confusion.
- Clear articulation of stop names
- Smooth transitions between route segments
- Reduced cognitive load for listeners
Calming tone for delays
When service is interrupted, Luna’s gentle delivery mitigates frustration. Her consistent pace helps maintain order and keeps passengers informed during wait times.
- Mitigates passenger frustration
- Consistent pacing during uncertainty
- Professional yet reassuring tone
Suitable for long announcements
Luna handles extended scripts effectively, maintaining clarity over longer durations. Her voice remains engaging without fatigue, making her suitable for comprehensive journey updates.
- Maintains clarity over duration
- Engaging without vocal fatigue
- Ideal for comprehensive updates
Aria: The warm narrator
Aria delivers a warm, clear narrative tone ideal for consistent fleet-wide messaging. As a female preset, she provides a steady, articulate voice that cuts through ambient transport noise without sounding robotic.
Clear and articulate
Her distinct enunciation ensures that route details and stop names are understood clearly, even in noisy stations. This clarity reduces passenger confusion and minimizes the need for repeated announcements.
Good for instructional updates
Aria excels at delivering step-by-step instructions, such as boarding procedures or transfer guidance. Her steady pace allows passengers to process information without feeling rushed.
- Ideal for multi-step boarding instructions
- Clear delivery of transfer points
- Effective for platform safety warnings
Professional yet friendly
She strikes a balance between corporate authority and approachability. This tone builds trust with passengers while maintaining the professional standard expected by large transport teams.
Nova: The bright announcer
Nova delivers a bright, upbeat tone ideal for high-energy transport environments. Her clear, articulate delivery ensures promotional messages cut through ambient noise without sounding robotic or overly formal. This voice profile maintains a consistent, professional yet friendly demeanor, making it suitable for time-sensitive announcements where listener attention is brief.
Upbeat and energetic
Nova’s energetic cadence keeps passengers engaged during short bursts of information. Her vocal style is designed to convey positivity and forward momentum, which aligns well with modern transit systems aiming for a welcoming atmosphere. The voice remains distinct and clear even at higher speeds, up to 4.0x, ensuring no loss in intelligibility during peak hours.
- Clear articulation for quick reading
- High energy without being shrill
- Consistent tone across languages
Good for promotional updates
Use Nova for announcing new routes, seasonal discounts, or loyalty program incentives. Her upbeat nature makes promotional content feel like an opportunity rather than a standard notice. This voice works effectively for digital signage in stations or audio blasts on platforms where you want to highlight positive service changes.
- Ideal for discount announcements
- Effective for new route launches
Engaging for short messages
Nova is suitable for messages under 100 characters. Her style prevents listener fatigue by delivering information with a crisp, lively finish. This makes her perfect for brief intercom updates, such as gate changes or immediate boarding calls, where clarity and tone must be immediate and memorable.
- Optimized for messages under 100 chars
- Reduces listener fatigue
- High retention for key info
Sage: The calm instructor
Sage delivers a calm, instructional tone ideal for guiding transport teams through standard operating procedures. Its steady pacing ensures clarity during routine broadcasts, making it a reliable choice for consistent fleet communication.
Ideal for safety instructions
The voice’s measured cadence reduces cognitive load for drivers receiving safety briefings. This clarity helps ensure critical guidelines are understood without distraction during complex maneuvers.
Slow and clear delivery
Sage operates effectively at reduced speeds, enhancing intelligibility in noisy cabin environments. This controlled delivery style prevents information overload during multi-step protocol announcements.
Good for emergency protocols
During unexpected events, Sage’s authoritative yet non-alarmist tone maintains order. It conveys urgency without panic, ensuring passengers and crew follow instructions precisely.
Felix: The energetic promoter
Felix delivers a high-energy voice profile ideal for short, attention-grabbing announcements. His tone is distinctively upbeat, making him suitable for promotional updates or service highlights where immediate passenger engagement is required.
Good for service promotions
Use Felix for brief messages announcing new routes or special transport offers. The energetic delivery captures attention quickly, ensuring passengers notice time-sensitive information without needing lengthy explanations.
- Ideal for short promotional blasts
- Effective for announcing new services
- Suitable for limited-time transport offers
High energy tone
Felix’s voice maintains a consistent, lively pace. This ensures that announcements remain clear and dynamic, even in noisy terminal environments or crowded platforms where lower energy voices might be overlooked.
Engaging for short blasts
For messages under 50 characters, Felix provides maximum impact. His crisp articulation ensures that key details like departure times or platform changes are understood instantly, reducing passenger confusion during peak hours.

Orion: The deep documentarian
Orion provides a deep, documentary-style voice suitable for historical or contextual fleet announcements. His tone is authoritative and steady, making him ideal for narrating route histories or station backgrounds.
Good for documentary-style updates
Orion’s vocal range offers a distinct narrative quality. Use him for pre-recorded messages that explain the history of a specific line or station, providing passengers with engaging context beyond simple scheduling data.
- Narrative delivery style
- Historical route information
- Station heritage announcements
Deep and authoritative
The deep timbre ensures clarity even in noisy environments. This characteristic supports authoritative delivery without sounding aggressive, maintaining a professional corporate standard for large transport teams.
- Clear projection in noise
- Professional tone
- Steady pacing
Suitable for historical info
Orion is specifically tuned for longer, information-dense texts. His steady pace helps listeners absorb detailed historical data, making him a reliable choice for educational transport updates.
- Detailed historical content
- Educational announcements
- Consistent delivery
Choosing based on language
Each voice in the Paisita Voices API supports all 23 available languages, including Arabic and English. This allows a single voice preset to handle diverse linguistic requirements without needing multiple audio files or complex routing logic.
Arabic voice compatibility
Arabic is fully supported by all eight voice presets. When sending requests, explicitly set the language field to ar to ensure correct pronunciation, as the API guesses Latin-script text as English by default.
English voice compatibility
English is the default language for Latin-script inputs if no language field is provided. All voices render English with high clarity, making it suitable for standard route announcements and safety instructions in English-speaking regions.
Multilingual fleet strategies
Deploy a consistent voice identity across mixed-language regions by using the same voice preset for all supported languages. For example, use Atlas for neutral announcements in English, Spanish, and German fleets simultaneously.
Manage character limits by splitting texts into parts of up to 4096 characters per request. This ensures reliable audio generation even for long, multilingual route descriptions.
Technical considerations
Selecting the correct audio format and speed ensures efficient playback across diverse transport infrastructure. Our API supports mp3, wav, opus, flac, and pcm, allowing you to match the codec requirements of your existing announcement systems.
Audio format selection
Use MP3 for general broadcasts to balance quality and file size. WAV or FLAC is suitable for high-fidelity local PA systems where bandwidth is not a constraint. PCM is ideal for raw audio processing pipelines.
- MP3: Best for web and mobile apps
- WAV/FLAC: Ideal for high-quality local speakers
- OPUS: Efficient for low-bandwidth transmission
- PCM: Raw data for custom processing
Speed adjustment tips
Adjust the speed parameter (0.25–4.0) to fit the pacing of your announcements. Standard speed is optimal for most use cases. Increase speed for short alerts to save time; decrease it for complex route information to ensure clarity.
- 0.25–1.0: For clear, detailed instructions
- 1.0: Standard pace for general announcements
- 1.5–4.0: For quick alerts and updates
Character limit management
Each request supports up to 4096 characters. For longer announcements, split the text into parts and join the resulting audio files. This ensures consistent quality and avoids truncation issues in your transport fleet's audio system.
- Max 4096 characters per request
- Split long texts manually or via script
- Join audio files for seamless playback
Implementation best practices
Efficient integration requires managing API constraints while maintaining audio quality across large fleets. By grouping text segments and handling failures gracefully, transport teams ensure consistent delivery of critical announcements without excessive API calls or wasted credits.
Batch processing strategies
Split long announcements into segments under 4,096 characters and join the resulting audio files. This approach respects the per-request character limit while keeping processing time predictable, especially important for detailed route information or lengthy safety protocols.
- Group related messages to minimize round-trip latency.
- Use the
speedparameter to adjust playback duration if needed. - Process segments sequentially to avoid queue contention during peak hours.
Error handling in announcements
Monitor response status codes to distinguish between temporary server load and permanent failures. Failed requests are free, so retrying immediately after a 5xx error is safe, but implement exponential backoff to respect the 60 requests per minute rate limit.
- Log character counts and response times for debugging.
- Retry on transient errors with a short delay.
- Fallback to a default voice if a specific voice fails.
Testing voice clarity
Validate audio output in the actual environment where it will play. Check for clarity at different volume levels and background noise conditions typical of transport hubs or vehicles. Use the provided curl example to generate test files quickly.
- Test with varying
speedsettings for optimal clarity. - Verify language detection matches the intended script.
- Ensure audio format (mp3, wav, etc.) is supported by playback hardware.
Cost optimization
Optimizing your API spend requires aligning voice selection with announcement duration and content density. Since pricing is strictly character-based, all voices have the same price point.
Efficient usage also depends on precise text management and smart account setup. By minimizing unnecessary characters and leveraging available trial credits, you can reduce overhead while maintaining consistent audio quality across your multilingual fleet announcements.
Choosing efficient voices
All voices share the same price point of $0.01 per 1,000 characters, so voice selection does not affect unit cost. Focus instead on delivery speed and clarity to ensure messages are concise. For example, Felix’s energetic tone may convey short blasts quickly, while Sage’s calm instruction suits detailed safety protocols without needing repetition.
- Use concise phrasing to reduce character count per request.
- Match voice energy to message length to avoid redundancy.
- All voices have identical pricing regardless of tone.
Minimizing character count
Spaces and punctuation count as characters, so trim whitespace and avoid redundant greetings. Break long texts into chunks under 4,096 characters to ensure reliable processing. Failed requests are free, so test short segments first to verify clarity before committing to full-length announcements for your fleet.
Avoid unnecessary words that do not add value to the announcement. For instance, replace 'Please be advised that the bus is arriving' with 'Bus arriving' to save characters while maintaining clarity.
- Count spaces and punctuation in your total character usage.
- Split texts into 4,096-character segments for reliability.
- Test short segments before full deployment.
Leveraging trial credit
Sign in with Google to receive $0.50 in trial credit, valid for 7 days and sufficient for approximately 50,000 characters. This allows you to test voice profiles and integration workflows without immediate payment. Accounts created with email and password do not receive this trial credit, so consider your sign-in method carefully.
Use the trial period to validate audio quality across different languages and voices. Once your system is stable, top up with USDT on the TRON network for ongoing usage. Balance does not expire, so you can accumulate credits for future high-volume announcement periods.
- Google sign-in grants $0.50 trial credit.
- Credit lasts 7 days and covers ~50,000 characters.
- Email/password accounts start without trial credit.
Future-proofing your fleet
Scaling your transport network requires consistent API performance. Our system processes requests sequentially, so batch splitting ensures reliable delivery for large fleets without complex infrastructure overhead.
Scalability of API usage
Each API key supports 60 requests per minute, which handles standard announcement volumes effectively. For larger deployments, distribute load across multiple keys or split texts into 4096-character segments to maintain steady throughput.
- Use batch processing for multiple stops
- Distribute load across multiple API keys if needed
Updating voice profiles
Voice presets remain stable, but you can switch any text-to-audio call to a different voice instantly by changing the voice parameter in your JSON body. No retraining or new contracts are required.
Issue a new API key anytime via the dashboard to invalidate the old one immediately, ensuring secure updates across your entire fleet without service interruption.
Monitoring usage trends
Track character consumption and request status in your dashboard to anticipate top-up needs. Since pricing is strictly per 1,000 characters, monitoring usage helps you optimize text length and avoid unexpected queues.
Failed requests incur no charge, so monitor error rates to identify text formatting issues or network problems affecting your announcement delivery.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Her voice remains engaging without fatigue, making her suitable for comprehensive journey updates.
23 languages
Any voice speaks any of these languages.
