Frequently asked questions
Which TTS API is best for dubbing?
VoxLocale is well-suited for dubbing, supporting 23 languages and 8 distinct voices. Our API offers character-based billing at $10 per million characters, ensuring predictable costs for high-volume pipelines without hidden fees or minimum spend requirements.
How do voices compare in quality?
We offer eight presets like Aria (warm narrator) and Felix (energetic promo). All voices are based on licensed recordings from LibriTTS-R and LibriVox. Each generated file carries an inaudible watermark marking it as synthetic speech, ensuring consistent quality across all 23 supported languages.
Is OpenAI SDK better than cURL?
Our endpoint matches the OpenAI speech endpoint shape exactly. You can use the official OpenAI SDK by changing the base_url to https://api.playyy9.com/v1, or use plain cURL. Both methods work identically, so choose the tool that fits your existing infrastructure best.
What is the best price model?
Our pricing is straightforward: $10 per million characters, including spaces and punctuation. Failed requests are free. There are no monthly plans, seats, or subscriptions. You can top up with USDT (TRC-20) for a minimum of $5, and your balance never expires. This character-based model ensures you only pay for what you use, making it ideal for scaling your localization workflow.
Which voices are most realistic?
All eight voices are synthetic and carry an inaudible watermark. Realism varies by use case: Aria offers warm narration, while Orion provides a deep, documentary tone. Every voice is based on licensed recordings from LibriTTS-R and LibriVox. We do not offer custom voice cloning, so you select from these eight preset personalities rather than creating unique voices.
How many languages do I need?
We support 23 languages, including English, Spanish, German, Japanese, and Hindi. Any voice can speak any of these languages. If you do not set the language parameter, we guess it from the script, defaulting to English for Latin text. Vietnamese and Thai are not supported.
Is 60 requests per minute enough?
The limit is 60 requests per minute per key, or 10 per minute before your first top-up. For long texts, split them into 4096-character parts. Requests process sequentially, so under load, a request may wait in a queue, but this ensures stable delivery.
Do I need FLAC or is MP3 enough?
You can choose from mp3, wav, opus, flac, or pcm. MP3 is sufficient for most dubbing workflows due to its small size. FLAC offers lossless quality if you need archival-grade audio. The choice depends on your storage and bandwidth requirements, not audio fidelity differences.
How do I handle long scripts?
Split your text into parts of up to 4096 characters per request. Send each part to the API and join the resulting audio files. This approach ensures reliable synthesis for scripts longer than the character limit, maintaining consistency across your entire dubbing project.
Is data privacy important?
Input text is used only to produce audio and is not used for training. Generated audio is returned in the response and not stored. We keep request metadata like time and status for billing purposes only, ensuring your script data remains private and secure.
Can I use this for commercial projects?
Yes, there are no restrictions on commercial use. You pay for what you use, with no minimum spend or seat limits. You can issue a new API key at any time, and the old one stops immediately, giving you full control over your commercial workflow.
How do I switch from another provider?
Our API endpoint matches the OpenAI speech endpoint shape. You can switch by updating your base_url to https://api.playyy9.com/v1 and using your VoxLocale key. No new SDKs are needed; just update your configuration and start sending requests to our reliable infrastructure.
Key Selection Criteria
Selecting the right TTS API for dubbing requires balancing audio fidelity with operational efficiency. You need voices that convey the correct tone for your localization projects without sounding robotic, while ensuring the platform supports the specific languages your clients demand.
Voice Quality
Our eight preset voices are built from licensed LibriTTS-R recordings, offering distinct tones from warm narrators to energetic promo styles. Each audio file includes an inaudible synthetic watermark, ensuring clear identification of AI-generated content for your clients.
- 8 distinct presets: aria, nova, luna, sage, felix, milo, atlas, orion
Language Support
The API handles 23 languages, covering major European and Asian markets. Any voice can speak any supported language, allowing you to maintain consistent audio quality across diverse dubbing pipelines without managing multiple voice models.
- Supports 23 languages including English, German, Japanese, and Hindi
Pricing Model
We charge $10 per million characters, meaning you pay $0.01 per 1,000 characters. This character-based model is predictable for budgeting, with failed requests refunded in full. Start with a free trial credit to test volume.
- $0.01 per 1,000 characters
- Failed requests are free
Available Voices
Choosing the right voice profile is critical for maintaining audience engagement across different dubbing projects. Our platform offers eight distinct presets, each tuned for specific tonal requirements in professional localization.
These voices are derived from licensed recordings and public-domain audiobook readings, ensuring high-quality synthetic speech that sounds natural and consistent across 23 languages.
Female Voices
Female presets cover a range of narrative styles, from warm narration to energetic promotions. Aria provides a clear, warm tone ideal for narration, while Nova offers a bright, upbeat delivery suitable for promos.
Luna excels in soft, storytelling contexts, and Sage delivers a calm, instructional voice perfect for e-learning or instructional videos. Each voice adapts seamlessly to the target language without losing its character.
- Aria: Warm, clear narrator
- Nova: Bright, upbeat
- Luna: Soft, storytelling
- Sage: Calm, instructional
Male Voices
Male voices are optimized for dynamic content delivery. Felix brings energy to promotional material, while Milo offers a friendly, conversational tone that builds trust with viewers.
Atlas provides a neutral, news-style delivery, and Orion adds depth for documentary narration. These voices maintain consistent character across all supported languages.
- Felix: Energetic, promo
- Milo: Friendly, conversational
- Atlas: Neutral, news
- Orion: Deep, documentary
Neutral Voices
While each voice has a distinct character, Atlas and Milo serve as versatile options for neutral content where specific emotional bias is unnecessary. Their clarity ensures that the translated message remains the focus.
All voices support the full range of 23 languages, allowing agencies to maintain audio consistency even when switching between languages within a single project pipeline.
Language Coverage
Selecting the right engine hinges on linguistic breadth. Our platform supports 23 distinct languages, ensuring your dubbing projects cover major global markets without juggling multiple vendors.
This breadth allows agencies to standardize their technical stack while maintaining high-quality localization across diverse regions.
European Languages
We cover the core European markets including English, German, French, Spanish, Italian, Portuguese, Dutch, Polish, Russian, and Scandinavian languages like Swedish, Danish, and Norwegian. Greek and Turkish are also fully supported, making this a robust solution for Western and Central European dubbing workflows.
Every voice can speak any of these languages. If you do not set the language parameter, the system guesses from the script, defaulting to English for Latin text.
- English, German, French, Spanish, Italian, Portuguese
- Dutch, Polish, Russian, Turkish
- Swedish, Danish, Norwegian, Greek
Asian Languages
Asian localization is simplified with support for Chinese, Japanese, Korean, Hindi, and Malay. This allows agencies to handle East and Southeast Asian projects using the same API endpoint and voice presets.
Note that Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, Bengali and others are not supported, so verify your target languages against our list before committing to a pipeline.
- Chinese, Japanese, Korean
- Hindi, Malay
Global Reach
With 23 languages in total, you can scale your dubbing pipeline significantly. Whether you are localizing a single video or an entire series, the consistent API shape keeps your integration simple.
Use the GET /v1/voices endpoint to confirm available languages and formats for your specific project needs.
- Total of 23 supported languages
- Consistent API across all regions
Technical Integration
Integration is straightforward because the API mirrors the OpenAI speech endpoint structure. By simply updating the base_url to https://api.playyy9.com/v1, your existing OpenAI SDKs work immediately without code refactoring. This compatibility extends to plain HTTP requests, allowing flexible implementation in any language.
The consistent JSON request shape means you can reuse familiar patterns for authentication and payload construction. We provide clear documentation to help you integrate quickly into your dubbing pipeline.
OpenAI SDK Support
Our endpoint POST https://api.playyy9.com/v1/audio/speech accepts the same request body as OpenAI. Update your base_url to point to our API, and standard OpenAI SDKs function seamlessly.
This approach simplifies testing and deployment for teams already familiar with OpenAI’s developer tools, reducing the learning curve for new integrations.
- - Compatible with official OpenAI SDKs
- - No custom client library required
- - Standard JSON payload structure
cURL Compatibility
For quick testing or server-side scripts, the API supports standard cURL commands. You can send a POST request with JSON headers to receive audio files directly.
This flexibility allows developers to integrate the TTS service into bash scripts, Python scripts, or custom automation tools without heavy dependencies.
Documentation Quality
We provide clear, example-driven documentation at /docs/. It includes ready-to-use cURL examples and explains all request fields like model, voice, and speed.
The documentation is specific to our API, ensuring you know exactly what parameters are supported and how to handle responses for your dubbing workflow.
Pricing Structures
Pricing structures vary significantly among TTS providers, but our character-based model offers greater precision than per-minute models. This approach ensures you pay only for the text processed, avoiding hidden costs for audio length variations caused by voice speed or pauses.
We do not offer subscription plans or per-minute billing. Instead, you top up your balance with USDT (TRC-20) and pay strictly per character generated, making costs predictable and scalable for any volume.
Per Character
Our pricing is straightforward: $10 per 1 million characters, or $0.01 per 1,000 characters. Spaces and punctuation count as characters, providing exact cost transparency. Failed requests are free, ensuring you never pay for errors.
- $0.01 per 1,000 characters
- Spaces and punctuation count
- Failed requests are free
Per Minute
Our character-based approach offers greater precision than per-minute models. Since audio duration varies by voice and speed settings, per-minute billing can be unpredictable. Our model aligns cost directly with input volume, which is more accurate for dubbing workflows where text length is known upfront.
Subscription Models
We do not offer monthly subscriptions or seats. There are no minimum spends or recurring fees. You simply top up your balance with USDT on the TRON network (TRC-20) and use credits as needed. This pay-as-you-go model fits agencies with fluctuating project volumes perfectly.
- No monthly subscriptions
- No minimum spend
- Balance does not expire
Performance Metrics
For localization agencies, understanding generation speed is critical for pipeline planning. Our TTS API processes text at roughly real-time speeds, generating approximately 25 characters of audio per second for longer texts. This consistency helps agencies estimate production timelines accurately without unexpected delays.
Generation Speed
Short phrases may take a few seconds to synthesize due to initialization overhead. However, long-form content scales efficiently, maintaining a steady output rate that matches spoken tempo. This predictable timing allows teams to batch process dubbing tasks effectively.
Rate Limits
Each API key supports 60 requests per minute once funded. New accounts start with a limit of 10 requests per minute until the first top-up is processed. This structure encourages steady usage patterns rather than bursty, unmanageable spikes.
- 60 requests/minute per key after top-up
- 10 requests/minute for new accounts
- One active key per account
Queue Handling
Requests are processed sequentially. Under heavy load, a request may wait in a queue before synthesis begins. This ensures fair resource distribution but means peak times might slightly increase total processing time for large batches.
Audio Formats
Choosing the right audio format is critical for dubbing workflows. Our API supports MP3, WAV, FLAC, OPUS, and PCM, allowing you to balance file size and quality based on your distribution needs. Each format is generated instantly via the standard response_format parameter.
Using uncompressed formats like WAV or FLAC ensures maximum fidelity for professional mastering, while MP3 offers efficient storage for web delivery. All formats support the full speed range of 0.25x to 4.0x, preserving audio integrity regardless of pacing adjustments.
MP3
MP3 is a common choice for web-based dubbing and streaming platforms. It provides a good balance between file size and audio quality, making it ideal for final delivery to clients or direct upload to video hosting services.
The format is widely supported across editing software and players, ensuring seamless integration into your existing localization pipeline.
- Ideal for web distribution
- Small file sizes
- Universal compatibility
WAV
WAV offers uncompressed, high-quality audio perfect for professional post-production. Use this format when you need pristine audio for further editing, mixing, or mastering in your dubbing studio.
Since it is uncompressed, WAV files are larger than MP3s, but they retain the full dynamic range and detail of the generated speech, ensuring no quality loss during subsequent processing.
- Uncompressed audio
- Highest fidelity
- Best for editing
FLAC
FLAC provides lossless compression, offering the quality of WAV with smaller file sizes. It is suitable for archiving high-quality voice-overs or for clients who prefer lossless formats without the bulk of uncompressed WAV files.
This format ensures that your audio remains pristine while reducing storage costs, making it a smart choice for long-term project archives or high-end dubbing deliverables.
- Lossless compression
- Smaller than WAV
- Archival quality

Data Privacy
We prioritize your creative assets by ensuring input text is used exclusively for audio synthesis. Once the audio file is generated and returned in the response, we do not store the original text or the resulting audio on our servers.
This strict ephemeral storage model means your scripts and final voice-overs remain yours, reducing data retention risks for sensitive dubbing projects.
Input Usage
Your input text is processed solely to produce the requested audio file. We do not repurpose your scripts for other purposes or retain them after delivery.
Storage Policy
Generated audio is returned directly in the API response and is not stored. Only request metadata—such as timestamp, character count, and status—is kept briefly for accurate billing purposes.
Training Data
Unlike many consumer-grade platforms, we do not use your input text to train our models. Your content remains private and does not contribute to our voice engine development.
Customer Support
VoxLocale provides support exclusively through email, eliminating phone wait times and ensuring your queries are logged for precise technical troubleshooting. We do not guarantee a strict SLA, but our team typically responds to technical inquiries within a few hours during business days.
Email Only
Our support model relies entirely on email communication. This approach allows our engineers to review code snippets and error logs thoroughly before responding, ensuring accurate solutions for integration issues without the noise of live chat or phone queues.
- No phone number to call at any hour
- Direct access to engineering support via ticketing
- No live chat widgets cluttering your workflow
Response Time
We do not guarantee a strict SLA, but our team typically responds to technical inquiries within a few hours during business days. For urgent billing or API key issues, we prioritize these tickets to minimize downtime in your dubbing pipeline.
Documentation
Our comprehensive documentation includes ready-to-use cURL examples, SDK configurations, and clear explanations of rate limits and character limits. We regularly update the docs to reflect changes in voice presets and supported languages, ensuring you have the exact specifications needed for seamless integration.
- Ready-to-use cURL examples
- SDK configuration guides
- Clear rate limit explanations
Common Pitfalls
Even with a robust pipeline, small oversights can halt production. Understanding the hard limits of the API prevents unexpected errors during bulk dubbing jobs.
By anticipating these common pitfalls, agencies can design more resilient workflows that handle edge cases gracefully without manual intervention.
Character Limits
The API enforces a strict 4096-character limit per request. If your source text exceeds this, you must split it into chunks and join the resulting audio files manually or via script.
Short phrases may take a few seconds to generate, while longer texts process at roughly real-time speed (25 characters per second). Under load, requests queue sequentially.
- Split text >4096 chars before sending.
- Expect queuing delays during high volume.
- Join audio files post-generation.
Language Guessing
If the language field is omitted, the API guesses based on script. Latin script defaults to English, which causes errors for other Latin-based languages like Dutch or French.
Always explicitly set the language ISO code for non-English Latin text to ensure correct voice selection and pronunciation.
- Default is English for Latin script.
- Explicitly set
languageparameter. - Avoids mispronunciation errors.
Speed Constraints
Speed ranges from 0.25 to 4.0. Extreme speeds may degrade quality. Rate limits cap at 60 requests per minute per key, or 10 before top-up.
Plan your integration to respect these limits, especially when processing large dubbing projects in parallel.
- Limit speed between 0.25–4.0.
- Max 60 requests/minute.
- Queue requests if overloaded.
Final Recommendations
Choosing the right engine depends on your agency’s scale and technical depth. Our API offers a straightforward path for teams of any size, balancing cost efficiency with reliable delivery for professional dubbing workflows.
The character-based pricing model eliminates guesswork, ensuring you only pay for what you generate. With 23 languages and 8 distinct voices, you can maintain consistent quality across diverse projects without hiring additional talent.
For Small Agencies
Start with the free trial to validate quality against your specific content. The $0.50 credit covers approximately 50,000 characters, allowing you to test voice suitability before committing to top-ups via USDT.
Use the OpenAI SDK with a simple base_url change to integrate quickly. The predictable per-character cost helps manage budgets for smaller projects without unexpected overages.
- Test with the 7-day free trial credit.
- Use standard OpenAI SDKs for easy integration.
- Pay only for successful requests.
For Large Enterprises
Scale your dubbing pipeline by splitting long texts into 4096-character chunks and joining the audio files. This ensures consistent quality even for lengthy scripts, leveraging our 60 requests per minute limit efficiently.
Maintain brand consistency across 23 languages using our 8 preset voices. Each voice supports all supported languages, allowing you to assign specific tones to different regions or character types without managing multiple providers.
- Split long texts into 4096-character parts.
- Leverage 60 requests per minute limit.
- Maintain consistency across 23 languages.
For Developers
Implement the API using the standard OpenAI SDK, changing only the base_url to https://api.playyy9.com/v1. This allows you to use familiar tools and libraries while accessing our specific voice models.
Handle errors gracefully by checking for failed requests, which are refunded automatically. Use the GET /v1/voices endpoint to dynamically select voices and languages based on your project requirements.
- Use standard OpenAI SDKs.
- Handle failed request refunds automatically.
- Dynamically select voices via API.
Next Steps
To begin your automation journey, create an account at /get-api-key/ using Google or your email. You will receive a trial credit of $0.50, valid for seven days, allowing you to test the API without immediate payment.
After securing your key, verify your access by generating a few sample audio files. Check the balance and usage metrics in your dashboard to ensure the integration is working correctly before scaling your dubbing pipeline.
Sign Up
Visit the /get-api-key/ page to register your agency account. Choose Google for instant access or use your email and password. Your account is specific to playyy9.com and grants immediate access to the API key dashboard.
- Google or email registration
- Instant API key generation
- $0.50 trial credit (7 days)
Test Voices
Use your trial credit to generate audio with all eight voice presets. Verify that voices like Aria, Nova, or Atlas meet your quality standards across your target languages. Check the inaudible watermark and audio clarity.
- Test all 8 voice presets
- Verify audio quality and formats
- Check language support accuracy
Integrate API
Update your SDK’s base_url to https://api.playyy9.com/v1 and use your Bearer token. Send POST requests to /v1/audio/speech to convert your localized scripts into audio files. Monitor your balance and rate limits via the dashboard.
- Set base_url to api.playyy9.com/v1
- Use Bearer token for auth
- Send POST requests for speech
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Unlike many consumer-grade platforms, we do not use your input text to train our models.
23 languages
Any voice speaks any of these languages.
Sources and further reading
- Wikipedia
- LibriTTS-R corpus (OpenSLR 141)
- TRON developer hub: TRC-20
- openai-python on GitHub
- Creative Commons Attribution 4.0
- openai-node on GitHub
- MDN: Web Speech API
- Tether (USDT)
- LibriTTS-R paper (arXiv 2305.18802)
- IETF RFC 7845: Ogg Opus
- LibriVox public-domain audiobooks
- W3C WAI: making audio and video accessible
