Choosing the Right TTS API for Indie Transport Apps

By VoxRoute editorial team · Updated 2026-11-18

This guide helps indie developers building in-vehicle announcements, bus info systems, and logistics dashboards choose the right text-to-speech API. We evaluate VoxRoute’s TTS API for transport use cases, focusing on reliability, licensing clarity, and simple integration with existing software stacks.

You will learn how to evaluate voice quality, manage character limits, and handle long announcements efficiently. The guide covers pricing models, rate limits, and technical integration options, providing concrete examples and numbers to help you make informed decisions for your transport app.

Choosing the Right TTS API for Indie Transport Apps

Frequently asked questions

What is the best TTS software for indie developers?

Yes. VoxRoute provides a straightforward API compatible with OpenAI SDKs, supporting eight licensed voices and 23 languages. Pricing is transparent at $10 per million characters, with no monthly plans or minimum spend. This makes it suitable for indie transport apps needing reliable speech synthesis without complex infrastructure.

How do I compare TTS software prices?

Compare costs per character rather than per minute. VoxRoute charges $10 per million characters, or $0.01 per 1,000 characters. Failed requests are free. Unlike subscription models, there are no monthly plans or seats. This pay-as-you-go structure suits variable traffic in transport apps, ensuring you only pay for successful synthesis.

Which TTS API is easiest to integrate?

VoxRoute is designed for easy integration by mimicking the OpenAI speech endpoint. You send a POST request to /v1/audio/speech with JSON body parameters. The official OpenAI SDKs work by changing the base_url to https://api.99dtpix.com/v1. This allows developers to reuse existing transport software stacks with minimal code changes.

Can I use TTS software for multi-language apps?

Yes, the API supports 23 languages including English, Spanish, German, Chinese, and Japanese. Any voice can speak any supported language. You can optionally set the ISO language code; otherwise, it guesses from the script. Note that Vietnamese, Indonesian, Thai, and Ukrainian are not supported, so verify your target languages before integrating.

What is the difference between tts-1 and tts-1-hd?

Both models use the same engine and have the same price of $10 per million characters. The only difference is the model name. For transport announcements, standard quality is usually sufficient. Choose tts-1-hd only if higher fidelity is required, but note that processing time remains roughly real-time for both.

Do I need to pay for every API call?

You pay for successful synthesis. Failed requests are free, and the charge is returned in full. You are billed based on the input text length, including spaces and punctuation. The rate is $10 per million characters. This ensures you only pay for usable audio files generated by the API.

How accurate is the language detection?

If the language field is omitted, the API guesses from the script. Latin script defaults to English. For non-English Latin text, explicitly setting the language code is recommended for accuracy. Detection is not perfect for mixed scripts, so providing the ISO code ensures the correct voice and pronunciation for your transport announcements.

Can I use the API for commercial projects?

Yes, the voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain audiobooks. Generated audio carries an inaudible watermark marking it as synthetic speech. There are no restrictions on commercial use, making it suitable for in-vehicle announcements, bus info systems, and logistics dashboards without additional licensing fees.

What is the maximum text length for one request?

The maximum input text length is 4096 characters per request. For longer texts, you must split them into parts of up to 4096 characters and join the resulting audio files. This limit applies to both tts-1 and tts-1-hd models. Ensure your transport software handles chunking and concatenation for longer announcements.

Is the API suitable for real-time announcements?

It processes roughly 25 characters per second on long texts. Short phrases take a few seconds. Requests are processed sequentially, so under load, a request may wait in a queue. For critical real-time needs, monitor rate limits (60 requests/minute) and implement retry logic to handle potential delays.

Do I need technical skills to use the API?

Basic technical skills are required to send POST requests with JSON bodies. You need to manage API keys and handle responses. However, using the OpenAI SDK simplifies integration. The documentation provides curl examples. If you can write a simple script to call an HTTP endpoint, you can integrate the API into your transport app.

Can I switch from another TTS provider easily?

Switching is easy if you use the OpenAI SDK, as the endpoint shape is identical. You only need to change the base_url to https://api.99dtpix.com/v1 and update your API key. Ensure your code handles the specific voice IDs and language codes supported by VoxRoute.

What happens if I exceed my rate limit?

The rate limit is 60 requests per minute per key. If you exceed this, subsequent requests may be delayed or rejected until the window resets. Before the first top-up, the limit is 10 requests per minute. Monitor your usage and implement exponential backoff in your transport software to handle rate limiting gracefully.

Is the API reliable for critical transport apps?

The API is reliable for standard use but offers no SLA or uptime guarantee. Requests are processed sequentially, which may cause delays under load. There is no enterprise contract or on-premise option. For critical systems, implement redundancy and queue management in your application layer to handle potential API latency or temporary outages.

Understanding TTS for Transport

Sound wave illustration for: Understanding TTS for Transport

Text-to-Speech (TTS) systems convert written text into synthetic audio files via a simple API call. For transport apps, this means replacing static text displays or pre-recorded loops with dynamic, on-demand audio announcements that reflect real-time data.

Transport systems require clear, consistent voice output for safety and clarity. Indie developers need a solution that integrates easily with existing dashboards without requiring complex infrastructure or large teams to maintain audio assets.

Text-to-Speech Basics

TTS APIs accept text input and return an audio file in formats like MP3 or WAV. The process is straightforward: send a POST request with your announcement text, and receive a ready-to-play audio stream. This eliminates the need to record and store thousands of individual audio files for every possible route or delay scenario.

Modern TTS engines use neural networks to generate natural-sounding speech. While not perfect, they provide sufficient clarity for public announcements. The key is selecting a service that balances audio quality with API simplicity and cost efficiency.

  • Converts text to audio files (MP3, WAV, OPUS, FLAC)
  • Single API call for dynamic audio generation
  • No need to manage large libraries of pre-recorded clips

Transport Industry Needs

Clarity is paramount. Announcements must be understood in noisy environments like bus stops or train stations. Voices should be neutral and easy to follow, avoiding excessive emotion or variation that might confuse passengers.

Reliability ensures that announcements are generated quickly and consistently. Downtime or delays in audio generation can disrupt the passenger experience, especially during peak hours or unexpected events.

  • Clear, intelligible speech for noisy environments
  • Consistent audio quality across all announcements
  • Support for multiple languages and dialects

Indie Developer Constraints

Indie developers often work with limited budgets and small teams. A per-character pricing model ensures you only pay for what you use, avoiding expensive monthly subscriptions. A free trial allows testing without upfront costs.

Simplicity is key. Using the OpenAI SDK or direct cURL requests means leveraging familiar tools. This reduces the learning curve and allows for quick integration into existing transport software stacks.

  • Pay-per-use pricing with a free trial
  • Easy integration using OpenAI SDK or cURL
  • Low overhead for small development teams

Key Features to Evaluate

Sound wave illustration for: Key Features to Evaluate

Selecting the right TTS API hinges on balancing voice clarity with technical simplicity. For transport applications, the audio must remain intelligible over engine noise or ambient station sounds. The available voices range from warm narrators to neutral news readers, allowing you to match the tone of your specific use case, whether it’s a friendly bus update or a formal train announcement.

Language support is equally critical. The API covers 23 major languages, ensuring broad coverage for international routes. However, it does not support Vietnamese, Indonesian, Thai, or Ukrainian, so you must verify compatibility with your target regions. Always set the language parameter explicitly for non-English Latin text to avoid incorrect voice selection.

Voice Quality and Variety

The eight voice presets offer distinct tones suitable for different announcement types. Options like Atlas provide a neutral, news-style delivery ideal for schedules, while Luna offers a soft, storytelling tone for station information. All voices are based on licensed recordings from LibriTTS-R, ensuring clear, consistent audio output without the need for complex customization.

Each voice can speak any of the 23 supported languages, regardless of the voice’s origin. This flexibility allows you to maintain a consistent audio brand across different language segments without switching voice models. The generated audio carries an inaudible watermark, marking it as synthetic speech, which helps in content identification.

  • Eight distinct voice presets for varied tones
  • All voices support all 23 languages
  • Licensed recordings from LibriTTS-R (CC BY 4.0)
  • Inaudible watermark in generated audio

Language Support Coverage

The API supports 23 languages, including English, Chinese, Spanish, and German. If the language parameter is omitted, the system guesses the language based on the script, defaulting to English for Latin text. For non-English Latin-script text, explicitly setting the language code ensures accurate pronunciation and avoids misidentification.

Languages not supported include Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, and Bengali. If your transport route serves these regions, you must plan for alternative solutions or verify if the supported languages cover your primary audience. Always test with local text samples to confirm clarity in noisy environments.

  • 23 supported languages
  • Automatic language guessing with Latin default
  • Explicit language code recommended for non-English
  • Unsupported: Vietnamese, Thai, Ukrainian, etc.

API Simplicity

The API follows a simple POST request structure, similar to the OpenAI speech endpoint. Send text via JSON, and receive an audio file in your chosen format. This simplicity reduces integration time, allowing indie developers to focus on app functionality rather than complex TTS logic. The same request shape works with official OpenAI SDKs by adjusting the base URL.

Authentication is handled via a single Bearer token, making it easy to manage keys. Rate limits are set at 60 requests per minute, which is sufficient for most indie applications. Failed requests do not incur charges, ensuring cost efficiency during testing and development phases.

  • POST request with JSON body
  • OpenAI SDK compatible
  • Single Bearer token auth
  • 60 requests per minute limit
Key Features to Evaluate

Pricing Models Compared

Sound wave illustration for: Pricing Models Compared

Pricing is strictly per-character at $10 per million, or $0.01 per 1,000 characters. Failed requests incur zero charge, ensuring you only pay for successful synthesis. This model eliminates the risk of wasted subscription fees during low-usage periods common in niche transport applications.

Per-Character vs. Subscription

Unlike monthly plans, there are no seats or minimum spends. You pay exactly for the volume generated. This linear cost structure simplifies budgeting for indie developers who build bus or train info systems with unpredictable usage spikes.

  • Zero cost for failed requests
  • No monthly commitment or seats
  • Linear pricing: $0.01 per 1,000 characters

Free Trial Availability

Signing in with Google provides $0.50 in trial credit, valid for 7 days. This covers approximately 50,000 characters, enough to test voice quality and API integration. Email-only accounts start without this trial credit.

Top-Up Flexibility

Top-ups require USDT on the TRON network (TRC-20) with a $5 minimum. Credits are automatic within minutes. Balances do not expire, and there are no invoices or VAT charges, keeping accounting simple for small teams.

  • USDT (TRC-20) only
  • Minimum $5 top-up
  • Balance never expires

Technical Integration Options

Sound wave illustration for: Technical Integration Options

Integrating the VoxRoute TTS API is straightforward for indie developers already familiar with standard REST patterns. The service mirrors the OpenAI speech endpoint structure, allowing teams to leverage existing infrastructure with minimal code changes.

This compatibility reduces the learning curve significantly. Developers can use the official OpenAI SDK or direct HTTP requests.

OpenAI SDK Compatibility

The API endpoint https://api.99dtpix.com/v1/audio/speech accepts the same JSON body shape as OpenAI’s speech endpoint. By simply changing the base_url to https://api.99dtpix.com/v1 and providing your tts-... key, the official OpenAI SDKs work immediately.

This means you can use familiar libraries like openai in Python or Node.js. You just need to update the configuration to point to our API domain. No custom client libraries are required, saving development time.

  • Use standard openai package with custom base_url
  • No need for a dedicated VoxRoute SDK
  • Supports both Python and JavaScript/Node.js ecosystems

cURL Direct Requests

For developers preferring lightweight solutions or those without a specific SDK dependency, direct HTTP POST requests are fully supported. This approach is ideal for embedded systems or simple backend scripts where overhead matters.

You can test the endpoint instantly using cURL. Send a JSON payload with model, input, voice, and response_format fields. The API returns the audio file directly in the response body or as a file attachment.

  • No external library dependencies required
  • Ideal for lightweight backend services
  • Easy to test with standard command-line tools

Rate Limit Impact

Each API key is limited to 60 requests per minute once your account is active. This limit applies to all endpoints, including voice listing and model checks. For new accounts, the limit is stricter at 10 requests per minute until the first top-up.

Under load, requests are processed sequentially. If you exceed the limit, you will receive a 429 error. Plan your announcement queues accordingly, especially for real-time bus or train announcements where timing is critical.

  • 60 requests/minute for active accounts
  • 10 requests/minute for new accounts
  • Requests processed sequentially, may queue under load

Voice Selection Strategies

Sound wave illustration for: Voice Selection Strategies

Selecting the correct voice model directly impacts passenger clarity and system professionalism. Transport announcements require distinct tonal qualities to convey urgency or routine information effectively without causing listener fatigue.

Neutral vs. Character Voices

Neutral voices like Atlas (news) or Sage (instructional) provide reliable clarity for standard route updates. Character voices such as Orion (documentary) or Felix (promo) add distinct personality but may distract during critical safety announcements.

  • Atlas: Neutral, ideal for standard news-style updates.
  • Orion: Deep, suitable for documentary-style narrations.
  • Felix: Energetic, best for promotional station announcements.

Gender and Tone Choices

Female voices like Aria (warm narrator) or Luna (soft storytelling) offer approachable tones for customer service messages. Male voices like Milo (friendly) or female voices like Nova (upbeat) provide clear delivery for dynamic digital dashboards.

  • Aria: Warm, clear narrator for general info.
  • Luna: Soft, good for storytelling segments.
  • Milo: Friendly, conversational tone for user guidance.

Audience Preferences

Match voice density to route complexity. Use calm tones for long-distance travel narratives and brighter tones for high-frequency urban shuttles. Test multiple presets with actual route scripts to ensure intelligibility in noisy vehicle environments.

  • Test voice clarity in typical vehicle noise levels.
  • Use calming tones for long-haul segments.
  • Opt for brighter tones for frequent stop announcements.

Handling Long Announcements

Sound wave illustration for: Handling Long Announcements

Transport announcements often exceed the 4096-character limit per request. Developers must split long texts into smaller chunks to ensure reliable synthesis. This approach prevents payload errors and maintains consistent audio quality across all segments.

Splitting should occur at natural sentence boundaries to avoid cutting words mid-syllable. This ensures that the final joined audio sounds coherent to passengers. Efficient splitting reduces latency and keeps the user experience smooth during real-time announcements.

Character Limits Explained

Each API call accepts up to 4096 characters, including spaces and punctuation. Exceeding this limit causes a request failure. For long bus route descriptions or train schedules, you must divide the input text before sending it to the endpoint.

Splitting Text Efficiently

Break text at sentence endings or logical pauses rather than arbitrary character counts. This preserves the natural rhythm of the speech. For example, split a long station list after each stop name. This method ensures that each segment synthesizes independently without losing context or tone continuity.

Audio Joining Techniques

After receiving multiple audio files, concatenate them in the correct order. Use a simple audio library or server-side tool to merge the files seamlessly. The synthetic watermark remains intact across all segments. This ensures consistent branding and licensing compliance for the entire announcement.

Language Support Deep Dive

Sound wave illustration for: Language Support Deep Dive

Transport apps serve diverse riders, making language support critical. VoxRoute provides 23 languages, ensuring announcements reach non-English speakers without complex integrations. Each voice supports all listed languages, simplifying deployment for multi-regional routes.

However, gaps exist. Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, and Bengali are not supported. Developers must verify target regions before committing to specific voice presets for their transit systems.

Supported Languages List

The API supports Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish. Any voice can speak any of these languages.

  • Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish

Unsupported Language Gaps

Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, Bengali, and all others not in the list are not supported. If your route serves these populations, you may need a secondary provider.

Language Guessing Logic

If the language field is omitted, the API guesses from the script. Latin text defaults to English. For non-English Latin-script text, explicitly set language to ensure correct pronunciation.

Language Support Deep Dive

Audio Format Considerations

Sound wave illustration for: Audio Format Considerations

Selecting the correct audio format balances file size, decoding speed, and playback quality for your transport interface. The API supports mp3, wav, opus, and flac via the response_format field, allowing you to optimize based on your specific hardware constraints and network conditions.

For most in-vehicle announcements, efficiency and broad compatibility are paramount. Choose formats that minimize bandwidth while ensuring crisp audio reproduction across various speaker systems without introducing significant decoding latency.

MP3 for Efficiency

MP3 is the default choice for minimizing data transfer and storage. Its compressed nature makes it ideal for over-the-air updates or low-bandwidth bus Wi-Fi networks. Most modern infotainment systems and microcontroller-based audio players handle MP3 decoding with minimal CPU overhead, ensuring quick playback initiation for time-critical announcements.

WAV for Quality

WAV provides uncompressed, lossless audio, ensuring the highest fidelity for voice output. This format is suitable for high-end station displays or when precise audio timing is critical. However, the larger file size increases memory usage and transfer time, which may impact performance on resource-constrained embedded devices.

OPUS and FLAC Options

OPUS is a supported format for audio files. FLAC provides lossless compression, reducing file size compared to WAV while preserving exact audio quality, useful for archival storage of generated announcements without re-encoding artifacts.

Data Privacy and Usage

Sound wave illustration for: Data Privacy and Usage

Transparency is critical for transport applications where passenger trust matters. Our TTS API processes input text exclusively to generate the required audio file. Once synthesis is complete, the audio is returned directly in the response and is not stored on our servers, ensuring that passenger announcements or dashboard data remain ephemeral.

Billing accuracy relies on minimal tracking. We retain only request metadata—timestamp, character count, and status—to verify charges and process refunds for failed requests. No personal identifiers or conversation histories are archived, keeping your data footprint small and compliant with basic privacy expectations.

Input Text Usage

Input text is used only to produce the audio and is not used for training. The engine processes the string, generates the waveform, and discards the source text immediately after delivery.

This ensures that route instructions or dynamic announcements are not repurposed for model improvement, keeping your content proprietary to your application.

  • Text is not used for training models

Audio Storage Policy

Generated audio is returned in the response and not stored. Each request is a transient transaction; once the audio stream or file is delivered to your client, it is gone from our infrastructure.

This reduces storage costs and privacy risks, ideal for real-time bus or train announcements where historical playback is unnecessary.

Billing Metadata

Request metadata (time, characters, status) is kept for billing. This minimal dataset allows us to verify that charges match the character count processed and to issue full refunds if synthesis fails.

No other usage patterns or text contents are retained for analytics, ensuring your billing data remains strictly financial.

  • Metadata kept for billing only

Scaling Your Solution

Sound wave illustration for: Scaling Your Solution

Scaling your transport TTS solution requires managing rate limits and queue behavior carefully. With 60 requests per minute per key, you must design your integration to avoid throttling during peak announcement times. Understanding these constraints ensures your in-vehicle or dashboard announcements play without disruptive delays.

Proper queue handling and load testing prevent bottlenecks when processing multiple vehicle routes simultaneously. By aligning your request volume with the API’s capabilities, you maintain consistent audio delivery for bus, train, or logistics updates.

Rate Limit Management

Each API key allows 60 requests per minute after initial top-up. Before the first top-up, the limit is 10 requests per minute. Plan your batch sizes accordingly to stay within these bounds.

Use a single active key for simplicity. If you need higher throughput, issue a new key to stop the old one immediately, allowing parallel processing with a fresh rate limit allocation.

  • 60 requests/minute per key after top-up
  • 10 requests/minute before first top-up
  • One active key per account

Queue Handling

Requests are processed sequentially. Under load, a request may wait in a queue before synthesis begins. Factor this latency into your user experience, especially for real-time announcements.

Implement retry logic with exponential backoff to handle temporary spikes. Avoid sending burst requests that exceed the per-minute limit, which could cause unnecessary queuing delays.

  • Sequential processing
  • Queue delays under load
  • Retry with backoff

Load Testing

Test your integration with realistic text volumes. Send multiple requests simulating peak usage to observe queue behavior and response times.

Monitor request metadata for billing accuracy and latency. Adjust your batching strategy based on observed queue depths to ensure smooth audio playback.

  • Simulate peak usage
  • Monitor queue depth
  • Adjust batching strategy

Licensing and Rights Management

Sound wave illustration for: Licensing and Rights Management

VoxRoute delivers reliable, license-cleared speech synthesis by leveraging recordings from the LibriTTS-R dataset. These voices are derived from LibriVox public-domain audiobook readings, ensuring that the underlying audio material is free of restrictive copyright barriers for commercial integration.

This foundation allows indie developers to integrate high-quality text-to-speech into transport apps without navigating complex licensing agreements. The result is a straightforward API that handles the legal groundwork, letting you focus on implementation.

Licensed Recordings Source

All eight voice presets—such as Aria, Nova, and Atlas—are based on licensed recordings from the LibriTTS-R dataset. This dataset consists of public-domain audiobook readings from LibriVox, which are available under the CC BY 4.0 license.

By building on this established open dataset, VoxRoute ensures that the core audio material is legally clear for commercial use. Developers can use the generated audio for in-vehicle announcements or dashboard notifications.

  • Source: LibriTTS-R dataset
  • Origin: LibriVox public-domain audiobooks
  • License: CC BY 4.0

Commercial Use Rights

The public-domain status of the source recordings means you can generate audio for commercial transport applications without additional royalty fees. Whether you are building a bus information system or a logistics dashboard, the generated speech is ready for public use.

There are no hidden restrictions on where or how you play the audio. The API returns the audio file directly, and you own the output for your specific use cases, making it ideal for indie developers with tight budgets.

Attribution Details

While the source material is CC BY 4.0, the API response includes an inaudible watermark to identify the speech as synthetic. This helps distinguish generated content without requiring visible attribution in your app’s interface.

You are not required to display specific credits for the voice actors in your transport app. However, maintaining good practice by noting the use of synthetic voice in your app’s settings is recommended.

Licensing and Rights Management

Final Recommendations for Indie Devs

Sound wave illustration for: Final Recommendations for Indie Devs

Selecting the right TTS API for your transport app requires balancing cost, reliability, and ease of integration. With VoxRoute, indie developers can deploy voice announcements for buses, trains, and logistics dashboards without managing complex infrastructure or negotiating enterprise contracts. The API’s straightforward structure and predictable pricing make it a pragmatic choice for small teams building critical audio features.

This guide highlights how to leverage VoxRoute’s features effectively, from voice selection to handling long announcements. By following these recommendations, you can ensure your app delivers clear, compliant, and timely speech synthesis that meets the specific demands of the transport industry.

Start with the Free Trial

Begin by signing in with Google to receive $0.50 of trial credit, valid for seven days. This provides approximately 50,000 characters of free synthesis, allowing you to test the API’s quality without immediate financial commitment. If you prefer a traditional account, note that email-signup accounts do not include this trial credit.

Use this credit to generate sample audio using different voices and formats. Verify that the audio quality meets your standards for in-vehicle announcements before committing to a top-up.

  • Sign in with Google to activate $0.50 credit
  • Credit valid for 7 days
  • Email-only accounts start with zero credit
  • No expiration on topped-up balance

Test with Real Text

Input realistic transport announcements, including station names, times, and delays, to evaluate voice clarity and language guessing accuracy. Remember that Latin-script text defaults to English, so explicitly set the language parameter for non-English announcements to avoid mispronunciation. Test edge cases like numbers, abbreviations, and mixed-language scripts.

Evaluate the tts-1 and tts-1-hd models to compare quality against the 4096-character limit. Ensure that splitting longer announcements into smaller chunks maintains audio continuity when joined.

  • Set language explicitly for non-English text
  • Test station names and times
  • Verify splitting logic for long texts
  • Compare tts-1 vs tts-1-hd

Monitor Usage Metrics

Track your character consumption and request status via the dashboard. Each request costs $0.01 per 1,000 characters, and failed requests are free, so monitor failure rates to optimize your integration. Keep an eye on the 60 requests-per-minute rate limit, especially if your app triggers multiple announcements simultaneously.

Use the API’s response metadata to log character counts and status codes. This helps you predict costs and identify potential bottlenecks in your transport software stack before they affect user experience.

  • Track character consumption
  • Monitor 60 req/min limit
  • Log request status codes
  • Predict costs accurately

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.

Avatar of the voice aria
ariafemale
Avatar of the voice nova
novafemale
Avatar of the voice luna
lunafemale
Avatar of the voice sage
sagefemale
Avatar of the voice felix
felixmale
Avatar of the voice milo
milomale
Avatar of the voice atlas
atlasmale
Avatar of the voice orion
orionmale

23 languages

Any voice speaks any of these languages.

Sources and further reading

  1. Wikipedia
  2. Python documentation
  3. LibriVox public-domain audiobooks
  4. openai-node on GitHub
  5. W3C WAI: making audio and video accessible
  6. Creative Commons Attribution 4.0
  7. curl documentation
  8. TRON developer hub: TRC-20
  9. OpenAI API reference: create speech
  10. MDN: Web Speech API
  11. LibriTTS-R paper (arXiv 2305.18802)
  12. IETF RFC 7845: Ogg Opus

More on this site