Frequently asked questions
What is the character limit per API request?
Each API request supports a maximum of 4096 characters of input text. For longer announcements, you must split the text into parts of up to 4096 characters and join the resulting audio files. This ensures reliable processing within the standard request limits.
Does the API support streaming audio?
No, the API does not support streaming, WebSocket connections, or real-time partial audio. It returns the complete audio file upon request completion. You receive the full audio response after the synthesis process finishes, which typically takes a few seconds for short phrases.
Which voices work best for bus announcements?
For transport announcements, atlas (neutral, news) and sage (calm, instructional) are ideal for clear information delivery. felix (energetic, promo) suits dynamic updates, while aria (warm, clear narrator) provides a professional tone. All eight voices support the 23 available languages for diverse routes.
Can I use the API for real-time passenger information?
The API is suitable for passenger information systems, but requests are processed sequentially and may queue under load, so it is not true real-time streaming. Requests are processed sequentially, so under load, a request may wait in a queue. Typical speed is roughly real-time (25 characters per second), making it viable for standard transit update workflows.
How long does synthesis take for long texts?
Synthesis takes roughly real-time, generating about 25 characters of text per second of audio. For long texts, you split them into 4096-character parts. Requests are processed one after another, so under heavy load, a request may wait in a queue before processing.
Do I need to set the language parameter manually?
The language is guessed from the script if not set, defaulting to English for Latin text. However, for non-English Latin-script text, you should set the language parameter manually. This ensures accurate pronunciation for the 23 supported languages, including Arabic, Chinese, and Japanese.
What happens if a synthesis request fails?
If synthesis fails, the charge is returned in full. Failed requests are free, so you are not billed for unsuccessful audio generations. This ensures you only pay for successfully processed audio files, maintaining cost efficiency for your transport announcement system.
Is the audio generated from AI or recordings?
Voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain audiobook readings. Every generated file carries an inaudible watermark marking it as synthetic speech. The engine uses these recordings to produce text-to-speech output via the API.
Can I change my API key if it’s compromised?
Yes, you can issue a new API key at any time. When a new key is issued, the old one stops working immediately. This allows for quick security responses in your transport software stack.
Does the API support SSML for formatting?
No, the API does not support SSML, phoneme dictionaries, or custom pronunciation lexicons. It accepts plain text input up to 4096 characters. You control pacing and tone primarily through voice selection and the speed parameter, which ranges from 0.25 to 4.0.
Are there monthly minimums for API usage?
No, there are no monthly plans, seats, or minimum spend requirements. You pay only for what you use at $0.01 per 1,000 characters. Balance does not expire, and you can top up with USDT on the TRON network whenever needed.
How are characters counted in the text?
Characters include spaces and punctuation. The price is $0.01 per 1,000 characters. This counting method applies to all input text sent to the API, ensuring transparent billing for your transport announcement volume without hidden fees for common formatting elements.
Can I use the audio commercially?
The source recordings are from LibriVox public-domain audiobook readings, and the API provides a simple commercial license. The audio files are returned in the response and not stored by the service. You can use them in your transport software without additional royalties.
First request
curl
curl https://api.99dtpix.com/v1/audio/speech \
-H "Authorization: Bearer $TTS_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"tts-1","voice":"aria","input":"Hello from the API.","response_format":"mp3"}' \
--output hello.mp3Why VoxRoute for Transport?
VoxRoute provides text-to-speech synthesis built for indie developers integrating voice into transport software. Our API delivers license-cleared speech from LibriTTS-R and LibriVox recordings, eliminating complex rights management for bus, train, or logistics announcements. Generated audio carries an inaudible watermark confirming its synthetic origin, ensuring clear provenance for compliance audits.
License-Cleared Voices
All voices are derived from licensed recordings (LibriTTS-R CC BY 4.0) and public-domain audiobook readings. This foundation ensures that your in-vehicle announcements are legally cleared for commercial use without additional royalty fees or attribution overhead.
Indie-First API Design
We designed the endpoint to match the OpenAI speech shape, allowing you to use standard SDKs by simply changing the base URL. Authentication uses a single Bearer token, and requests return audio files directly via JSON. This simplicity reduces integration time for small engineering teams managing existing transport stacks.
23 Languages for Global Fleets
Support covers major transport hubs across Europe, Asia, and the Americas, including Arabic, Chinese, Hindi, Japanese, Korean, Spanish, and Russian. Non-Latin scripts are handled explicitly; if the language code is omitted, Latin text defaults to English. This ensures accurate pronunciation for diverse passenger demographics without requiring separate voice models.
How the API Works
Send text to the JSON endpoint via POST requests. The API accepts standard parameters and returns an audio file directly in the response body.
JSON Endpoint Integration
Point your HTTP client to the speech endpoint. Include your Bearer token in the Authorization header and send the text payload as JSON. The service processes the request and returns the audio file, ready for immediate playback in your transport software.
- Standard JSON request body
- Bearer token authentication
- Direct audio file response
Simple Voice Selection
Choose from eight preset voices optimized for clarity. Select a voice ID that matches your announcement style, such as a neutral tone for train platforms or a warm narrator for bus stops. The system handles the synthesis based on your selection.
Audio Format Options
Specify the output format in your request. You can choose between MP3, WAV, OPUS, FLAC, or PCM depending on your playback system requirements. This flexibility ensures compatibility with various in-vehicle audio hardware and dashboard displays.
- MP3, WAV, OPUS, FLAC, PCM support
- Format selection via request parameter
- Optimized for transport hardware

Voices for Announcements
VoxRoute provides eight distinct voice presets, each optimized for specific announcement contexts within transport systems. These voices are derived from licensed recordings and public-domain audiobook readings, ensuring clear, reliable audio for passenger information and driver alerts.
Selecting the right voice tone enhances communication clarity. Below are three presets commonly used in transport applications, described by their audible characteristics rather than subjective emotional impact.
Aria: Warm Narrator
Aria delivers a warm, clear tone. It is suitable for station arrival messages. Whether it reduces passenger anxiety is subjective and not guaranteed by the API.
Atlas: Neutral News Style
Atlas provides a neutral, news-style delivery. It is suitable for safety warnings and schedule updates. Its ability to cut through ambient noise depends on the playback volume and environment.
Milo: Friendly Conversational
Milo offers a friendly, conversational style. It is suitable for driver alerts or casual route information.
Technical Specifications
VoxRoute processes requests sequentially, which means under load a request may wait in a queue. This linear processing ensures consistent output but requires planning for high-frequency announcement systems.
Rate Limits & Queuing
The API allows 60 requests per minute per key. Before the first top-up, the limit is 10 requests per minute. Requests are processed one after another, so queuing may occur during peak usage.
- 60 requests/minute after top-up
- 10 requests/minute before first top-up
- Sequential processing with potential queuing
Character Limits per Request
Each request accepts up to 4096 characters. For longer announcements, split the text into parts and join the resulting audio files. Spaces and punctuation count toward the character limit.
- Max 4096 characters per request
- Split and join for longer texts
- Spaces and punctuation count as characters
Speed Adjustment Range
Adjust playback speed between 0.25 and 4.0 times normal rate. Typical generation speed is roughly real-time, producing about 25 characters of text per second of audio.
- Speed range: 0.25–4.0
- Real-time generation speed
- Approx. 25 chars/sec
Pricing Transparency
VoxRoute charges $0.01 per 1,000 characters, scaling to $10 per million characters. Every character, including spaces and punctuation, counts toward this flat rate. Failed requests incur no charge, ensuring you pay only for successful synthesis.
Cost per Million Characters
The pricing model is straightforward: $10 for every million characters processed. This applies uniformly across both tts-1 and tts-1-hd models, which share the same engine and pricing tier.
- $10 per 1,000,000 characters
- $0.01 per 1,000 characters
- Spaces and punctuation count as characters
- Failed requests are free
Free Trial Credit
New accounts signing in with Google receive a $0.50 trial credit, valid for 7 days. This provides approximately 50,000 characters for testing integration. Accounts created via email and password start without this trial credit.
USDT Top-Up Only
We do not accept credit cards, PayPal, or subscriptions. All top-ups are made via USDT on the TRON network (TRC-20). The minimum deposit is $5, and funds are credited automatically within minutes.
- USDT (TRC-20) only
- Minimum top-up: $5
- Automatic credit within minutes
- No VAT invoices or subscriptions

Use Cases in Transport
VoxRoute’s API handles the repetitive, high-volume text-to-speech tasks common in transport systems. By splitting long scripts into 4096-character chunks, the service ensures clear audio for stop announcements and platform messages without complex integration.
Bus Stop Announcements
Generate precise arrival and departure messages for digital displays or audio speakers. The API supports 23 languages, allowing fleets in diverse regions to deliver accurate local announcements without managing separate voice assets.
- Supports Arabic, Chinese, French, German, Spanish, and 19 other languages
- Automatic language detection for Latin script, with manual override via
languagefield - Consistent voice output across all stops using presets like Aria or Atlas
Train Platform Messages
Deliver real-time updates for delays, gate changes, or track shifts. With a rate limit of 60 requests per minute, the system can handle bursty traffic during peak hours, queuing requests if necessary.
- Typical generation speed: roughly 25 characters per second for long texts; short phrases may take a few seconds.
- Reliable fallback to standard formats (mp3, wav, opus) for legacy speakers
- No SSML needed; plain text input integrates directly with dispatch software
Fleet Driver Alerts
Send concise alerts to driver dashboards or mobile apps. The low-cost structure ($0.01 per 1,000 characters) makes frequent, short alerts affordable for large fleets. Audio is returned instantly in the response.
- Cost-effective for high-volume, short-duration messages
- Works with existing transport software stacks via standard HTTP POST
- Inaudible watermark confirms synthetic origin for compliance
Language Support
VoxRoute supports 23 languages, ensuring your transport announcements reach the right audience regardless of the region. The API automatically detects the language from the text script, but you can explicitly set the ISO code to ensure accuracy, especially for Latin-script texts that might otherwise default to English.
Latin Script Defaulting
By default, Latin-script input is processed as English. If your announcements are in French, German, or Spanish, you must set the language parameter to prevent mispronunciation. This explicit configuration guarantees that the synthetic voice adapts correctly to the specific phonetic rules of your target language.
- Latin script defaults to English without explicit configuration
- Set the
languageparameter for non-English Latin texts - Supported: Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish
Non-Latin Script Support
The API handles non-Latin scripts such as Chinese, Japanese, Korean, and Arabic natively. The language parameter can be used to specify the exact voice behavior for these scripts, ensuring correct tone and pacing for transport announcements in diverse global fleets.
Unsupported Languages
Be aware that the following languages are not yet supported: Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, Bengali, and others. If your fleet operates in these regions, you may need to use English or a supported alternative for clarity.
- Vietnamese
- Indonesian
- Thai
- Ukrainian
- Romanian
- Czech
Data & Privacy
VoxRoute uses your input text strictly for the purpose of generating the requested audio file. We do not use your input text for model training or to improve the underlying speech engine.
No Training on Input
Your input text is processed solely to produce the audio output and is not retained for training purposes. This ensures that proprietary route announcements or specific fleet data remain private to your application logic.
No Audio Storage
Generated audio files are returned directly in the API response and are not stored on our servers after delivery. Each request generates a fresh audio file without retaining the output for future reference or analysis.
Metadata for Billing
We retain request metadata including timestamp, character count, and status code to facilitate accurate billing and usage tracking. This minimal data allows us to process your USDT top-ups and ensure correct charge calculations based on the $10 per million characters rate.

Getting Started
Access the /get-api-key/ page to create an account using Google or an email address. The dashboard immediately displays your active API key, current balance, and a ready-to-use cURL example for testing the endpoint.
Sign Up for API Key
Accounts are tied to this specific site. Once registered, you receive one active key. If needed, you can issue a new key at any time, which invalidates the previous one instantly.
Try the Free Trial
Signing in with Google grants $0.50 of credit, valid for 7 days. This covers approximately 50,000 characters. Email-based accounts start without trial credit.
View Documentation
Check the /docs/ page for the full JSON schema, rate limits (60 requests/minute), and supported parameters like speed and format.
- Base URL: https://api.99dtpix.com/v1
- Endpoint: POST /v1/audio/speech
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Welcome to the developer hub, where your creative vision meets the stage.
23 languages
Any voice speaks any of these languages.
