Frequently asked questions
What is TTS?
Text-to-speech is a cloud API that converts text into audio files. You send a POST request with your script, and receive synthesized speech. It supports 23 languages and 8 distinct voice presets, allowing localization agencies to automate voice-over production without hiring extra talent for every project.
How does TTS work for dubbing?
Split your script into parts under 4096 characters, send them via API, and join the resulting audio files. The system handles 23 languages, letting you scale your dubbing pipeline. Failed requests are free, and you only pay for successful character generation, streamlining your production workflow.
Is TTS audio realistic?
Voices are based on licensed recordings from LibriTTS-R and LibriVox. Each generated file carries an inaudible watermark marking it as synthetic speech. While realistic, it is clearly machine-generated. We do not offer custom voice cloning or emotion tags, so the audio remains distinct from human recordings.
Can TTS replace human voice actors?
It replaces them for volume and speed, not necessarily for unique artistic direction. You have 8 presets like Aria or Atlas. For complex emotional nuance, humans are still needed. However, for scalable dubbing across 23 languages, TTS offers a cost-effective, consistent alternative to hiring multiple actors.
What languages does TTS support?
We support 23 languages: 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, ensuring consistent quality across your entire localization pipeline.
How accurate is the pronunciation?
Pronunciation depends on the voice model trained on public-domain audiobooks. If the language is not detected, set the language field explicitly for non-English Latin text. The API guesses from script otherwise. Accuracy is high for supported languages but relies on standard text inputs without custom lexicons.
Can I adjust the voice speed?
Yes, you can adjust speed between 0.25 and 4.0 using the speed parameter. This allows you to match pacing for different video styles. The API processes requests sequentially, so adjust speed to optimize generation time while maintaining the desired tone for your localized content.
Is the audio royalty-free?
The voices are based on licensed recordings (LibriTTS-R) and public-domain readings (LibriVox). Generated audio is returned in the response and not stored. You pay per character used. While the base recordings are public domain, the synthetic output includes an inaudible watermark for identification purposes.
Do I need technical skills to use TTS?
Basic API knowledge is helpful but not required. Our endpoint matches the OpenAI speech format, so existing OpenAI SDKs work by changing the base URL. We provide a ready curl example on the dashboard. You can also use plain HTTP requests to integrate TTS into your workflow.
How long does generation take?
Generation is roughly real-time, about 25 characters per second. Short phrases take a few seconds. Requests are processed sequentially, so under load, you may wait in a queue. For long texts, split them into parts to manage processing time and ensure consistent delivery.
Can I use TTS for YouTube videos?
Yes, you can generate audio for YouTube videos. The API returns standard audio formats like mp3, wav, opus, flac, or pcm. You can adjust speed and choose from 8 voices. Just ensure your script is under 4096 characters per request to fit our API limits.
Is TTS suitable for commercials?
Yes, voices like Felix offer an energetic, promo style suitable for commercials. You can adjust speed up to 4.0 for dynamic pacing. Since voices are based on licensed recordings, the audio is professional. Note that no emotion tags are available, so tone is fixed to the preset.
How do I choose the right voice?
Choose from 8 presets: Aria, Nova, Luna, Sage, Felix, Milo, Atlas, or Orion. Each has a distinct tone, from warm narrator to deep documentary. Use the /v1/voices endpoint to check details. Select the voice that best matches your content’s mood and target audience language.
What is the character limit?
Each request supports up to 4096 characters. Spaces and punctuation count toward this limit. For longer texts, split them into parts and join the resulting audio files. This ensures consistent generation quality and prevents timeouts, especially when processing large volumes of dubbing scripts.
Defining Text-to-Speech in Dubbing
Text-to-Speech (TTS) converts written scripts into spoken audio, enabling localization agencies to generate voice-overs without booking talent for every project. By sending text to an API, you receive ready-to-sync audio files, streamlining the path from script to final dub.
This approach replaces manual recording sessions with automated generation, allowing teams to produce content at scale. It is a foundational tool for agencies managing high volumes of localized video content across multiple markets.
What is TTS?
TTS is a cloud API that takes input text and returns an audio file. It is not a standalone app but an infrastructure component you integrate into your workflow. You send a JSON payload, and the system returns synthesized speech in seconds.
Synthetic Voice Generation
Our voices are synthetic, derived from licensed recordings of human actors. They are not direct recordings but engineered presets that simulate natural speech patterns. Each voice, from the warm narrator 'Aria' to the deep documentary voice 'Orion', offers a distinct tone suitable for different dubbing needs.
- Voices are based on licensed recordings from LibriTTS-R (CC BY 4.0)
- Each file contains an inaudible watermark marking it as synthetic speech
Automated Audio Production
Automation removes the bottleneck of scheduling voice actors. You define the voice, language, and speed, then receive the audio file directly. This allows for rapid iteration: change a line in the script, regenerate the audio, and replace the track in your video editor instantly.
Why Agencies Use TTS
Agencies adopting TTS replace expensive voice talent fees with a predictable, character-based pricing model. At $0.01 per 1,000 characters, you can generate hours of dialogue for the cost of a single session, eliminating per-minute rates and minimum booking fees.
This shift allows teams to allocate budget toward post-production or new projects. With no monthly subscriptions or minimum spend, you only pay for what you actually synthesize, keeping overhead low even during fluctuating workload periods.
Cost Efficiency
Agencies replace expensive voice talent fees with a predictable, character-based pricing model. At $0.01 per 1,000 characters, you can generate hours of dialogue for the cost of a single session, eliminating per-minute rates and minimum booking fees.
This shift allows teams to allocate budget toward post-production. With no monthly subscriptions, you only pay for what you synthesize, keeping overhead low.
- $0.01 per 1,000 characters
- No per-minute talent fees
- No minimum spend
Rapid Prototyping
Use synthetic voices to create quick audio drafts for client approval before booking human actors. This reduces revision cycles and ensures the script direction is locked in before committing to production budgets.
Generate multiple tone variations instantly using different presets like 'warm' or 'energetic' to find the perfect fit without scheduling multiple recording sessions.
- Quick client approval
- Tone exploration
Scalability
Handle large-scale dubbing pipelines for 23 languages without managing multiple freelance contractors. Our API processes requests sequentially, allowing you to generate consistent audio for hundreds of videos simultaneously.
With 8 distinct voices available, you can maintain character consistency across diverse projects while scaling your output volume without hiring additional talent.
- 23 languages supported
- 8 voice presets
- No talent bottlenecks
How It Works
Once your script is ready, the API accepts text via a simple POST request. You can send up to 4,096 characters per call, making it easy to handle individual scenes or dialogue lines without complex chunking logic for short clips.
Choose from eight distinct presets like Aria or Atlas. Each voice handles all 23 supported languages natively. You can also adjust speed between 0.25x and 4.0x to match the pacing of your video visuals perfectly.
The endpoint returns audio files in formats like MP3, WAV, or Opus. Generated files include an inaudible watermark to identify synthetic speech. For longer texts, split your script into 4,096-character parts and join the resulting audio files seamlessly.
Text Input
Send your localized script via JSON. The API handles up to 4,096 characters per request. Spaces and punctuation count toward this limit, so ensure your character count stays within bounds for smooth processing.
- Max 4,096 characters per request
- JSON body format
- Language auto-guess or explicit ISO code
Voice Selection
Pick from eight presets tailored for different tones, from warm narration to energetic promos. All voices support all 23 languages, ensuring consistent quality across your entire dubbing pipeline without hiring new talent for each language.
- 8 distinct voice presets
- Supports all 23 languages
- Adjustable speed (0.25–4.0x)
Audio Output
Receive high-quality audio files directly in your response. Choose formats like MP3, WAV, or FLAC. Each file carries an inaudible watermark marking it as synthetic speech, helping you track your generated content efficiently.
For longer scripts, split text into 4,096-character chunks and join the resulting audio files. This allows you to maintain precise control over timing and synchronization with your video timeline.
- Formats: MP3, WAV, Opus, FLAC, PCM
- Inaudible synthetic watermark included
- Support for multi-part audio joining
Benefits for Localization
For localization agencies, our TTS API handles 23 languages including Arabic, Chinese, and Swahili. Any of our 8 voice presets can speak any supported language, allowing you to dub content without hiring separate talent for each region. This flexibility lets you scale your video dubbing pipeline efficiently.
Multi-Language Support
Our API supports 23 languages, from English and Spanish to Hindi and Turkish. Any of our 8 voices can perform in any supported language, letting you maintain a consistent brand voice across diverse markets. If the language is not specified, the system guesses from the script, though explicit codes are recommended for non-English Latin text.
- Supports 23 languages including 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 supported language.
Consistent Tone
Maintain a uniform auditory identity across your entire library. Whether you choose the warm narrator 'Aria' or the calm instructor 'Sage', the same voice model applies globally. This consistency ensures your brand sounds professional and recognizable, regardless of the target language.
Instant Delivery
Receive audio files directly via the API response. While longer texts process in roughly real time, short phrases are generated quickly. Failed requests are free, ensuring you only pay for successful, usable audio files that integrate seamlessly into your editing workflow.
Limitations to Know
While our TTS API delivers consistent, high-quality narration, it operates as a precise engine rather than an emotional actor. Understanding these constraints helps you design workflows that leverage speed without sacrificing intelligibility or tone consistency.
No Emotion Tags
The engine does not support explicit emotion tags or dynamic inflection controls. Each voice preset (e.g., "warm" Aria or "deep" Orion) has a fixed tonal character. You cannot instruct the API to sound "angry" or "excited" in real-time; the tone is baked into the voice model itself.
No Custom Cloning
We do not offer voice cloning or custom voice design. You must choose from the eight available presets based on licensed recordings. If your project requires a specific celebrity or brand voice, it must match one of our existing models, as you cannot upload a reference audio file to create a unique clone.
Character Limits
Each API request accepts a maximum of 4,096 characters. For longer scripts, you must split the text into chunks and join the resulting audio files. This requires basic scripting logic but ensures reliable processing without time-outs.
Technical Requirements
Integrating VoxLocale into your dubbing pipeline is straightforward. The API endpoint follows the standard OpenAI shape, allowing you to send POST requests with JSON bodies containing your text, selected voice, and desired output format. Authentication relies on a single Bearer token, which you can regenerate instantly if needed.
API Integration
You can call POST https://api.playyy9.com/v1/audio/speech directly or use any standard HTTP client. The system processes requests sequentially, so expect slight queuing during high load. Each request accepts up to 4096 characters of input text.
- Endpoint:
POST /v1/audio/speech - Auth:
Authorization: Bearer tts-... - Rate limit: 60 requests/minute
SDK Compatibility
While we do not offer a proprietary SDK, the request structure mirrors the OpenAI speech endpoint. This means you can use the official OpenAI SDKs by simply changing the base_url to https://api.playyy9.com/v1 and providing your VoxLocale key.
Format Compatibility
Select your preferred audio format in the request body: mp3, wav, opus, flac, or pcm. This flexibility ensures seamless integration with most video editing software and dubbing workflows without requiring immediate post-processing conversion.

Use Cases
Text-to-speech API integration allows localization agencies to replace manual voice-over recording with automated audio generation. By sending script text to the endpoint, teams receive ready-to-sync audio files in formats like MP3 or WAV, fitting directly into existing editing timelines without scheduling talent.
This approach eliminates the logistical bottleneck of coordinating voice actors across multiple time zones. Agencies can generate multiple language versions of the same explainer or training video simultaneously, ensuring consistent tone across all localized assets while maintaining tight production schedules.
Explainer Videos
Agencies producing marketing or product demo videos can rapidly generate voice-overs in 23 languages. With eight distinct voice presets like Aria or Felix, teams can match the desired brand personality without re-hiring talent for each market. The API handles character limits up to 4096 per request, allowing scripts to be split and stitched seamlessly.
This enables quick A/B testing of different voice styles or scripts. If a script changes, the audio regenerates in seconds, keeping the production pipeline fluid and cost-effective for high-volume explainer content.
- Generate audio in 23 languages simultaneously
- Choose from 8 distinct voice presets
- Rapid iteration for A/B testing scripts
- No talent scheduling required
E-Learning
For educational content, consistency is key. The same voice can be used across all modules in a course, whether in English, Spanish, or Japanese. The API’s deterministic nature ensures that pronunciation and pacing remain uniform, which is critical for learner retention and comprehension across different language versions.
Educational agencies can update course materials by simply re-running the text through the API. This reduces the cost of maintaining multi-language courses, as changes to the script result in immediate audio updates without re-recording sessions.
- Consistent voice across all course modules
- Support for 23 languages including Hindi and Arabic
- Instant updates for script revisions
- Cost-effective scaling for large libraries
News Updates
News agencies require neutral, clear voices for daily updates. The Atlas or Milo presets provide a professional, conversational tone suitable for delivering factual content. With the API, agencies can produce localized news briefs in multiple languages quickly, ensuring timely distribution across global markets.
The ability to specify speed (0.25–4.0) allows for adjusting the pace to match video footage or audio constraints. This flexibility ensures that the generated speech fits precisely within video timelines, reducing post-production editing time.
- Neutral voices like Atlas for news
- Adjustable speed (0.25–4.0)
- Quick generation for daily updates
- Support for 23 languages
Comparison to Human Voice-Over
Switching from human voice-overs to our TTS API changes the economics and timeline of your dubbing projects. While humans record session by session, our API generates audio on demand, allowing agencies to handle volume spikes without scheduling delays or renegotiating rates with individual talent.
Speed
Human voice-over requires casting, scheduling, recording, and editing, often taking days per language. Our API generates audio in seconds. For long texts, synthesis runs at roughly real-time speed (25 characters per second), and short phrases take only a few seconds. This allows you to deliver dubbed content instantly without waiting for studio availability.
- Generation takes seconds, not days
- Sequential processing may increase turnaround time under load
- No waiting for actor availability
Cost
Human voice-overs involve per-project fees, session charges, and potential revision costs. Our pricing is strictly usage-based: $10 per million characters ($0.01 per 1,000 characters). There are no monthly plans, no seat licenses, and no minimum spend. Failed requests are free, so you only pay for successful audio.
- $0.01 per 1,000 characters
- No per-session or studio fees
- Pay only for successful requests
Quality
Our voices are based on licensed recordings from LibriTTS-R and LibriVox, providing clear, natural-sounding speech in 23 languages. While humans offer nuanced emotional interpretation, our eight presets (e.g., Aria, Atlas, Orion) cover narrator, news, and documentary tones effectively. Synthetic audio includes an inaudible watermark, distinguishing it from human recordings.
- 8 distinct voice presets
- Supports 23 languages
- Licensed, clear synthetic audio
Getting Started
Start by visiting /get-api-key/ and signing up with Google or email. This creates your agency account on playyy9.com, where you can view your initial balance and usage metrics immediately.
Sign Up
Google sign-in includes a $0.50 free trial valid for 7 days, perfect for testing dubbing pipelines without immediate payment. Email-based accounts start with zero balance but offer the same full API access.
Generate Key
Your dashboard displays a unique tts-... API key. This key authenticates your requests to https://api.playyy9.com/v1. You can regenerate it anytime to revoke access instantly, ensuring secure multi-agent workflows.
- One active key per account
- Instant revocation on regeneration
- Bearer token format
Test API
Use the provided curl example or OpenAI SDK to send a POST request to /v1/audio/speech. Pass your key, a voice preset like aria, and up to 4096 characters of text to receive an audio file.
- Supports mp3, wav, opus, flac, pcm
- Rate limit: 60 requests/minute
- Base URL: https://api.playyy9.com/v1
Common Questions
TTS generates synthetic speech that sounds natural for most professional contexts, though it lacks the nuanced emotional range of a human performer. Every file carries an inaudible watermark confirming it is AI-generated.
Audio files are returned in standard formats (mp3, wav, opus, flac, pcm), allowing you to import them directly into your editing timeline for fine-tuning timing or adding background music.
Is it realistic?
The voices are based on licensed recordings from LibriVox and are designed to sound clear and professional. While they lack complex emotion tags, they handle 23 languages with high accuracy. The output is suitable for explainer videos, news updates, and e-learning modules where clarity is key.
- Based on licensed LibriTTS-R recordings
- Inaudible synthetic watermark included
- Supports 23 languages
Can I edit the audio?
Yes. You receive a standard audio file, not a proprietary format. You can cut, trim, or process the audio in any DAW or video editor. Since the API returns the full audio file, you have complete control over the final mix.
- Formats: mp3, wav, opus, flac, pcm
- Compatible with any standard editor
- No proprietary format lock-in
Is it commercial?
The generated audio is yours to use commercially. There are no recurring licensing fees or per-view royalties. You pay only for the characters generated. The voices are based on licensed recordings (LibriTTS-R) and public-domain readings (LibriVox).
- No per-view royalties
- No recurring licensing fees
- Source: Licensed + Public Domain
Resources
Access comprehensive guides to integrate our TTS API into your localization pipeline. We provide detailed references for endpoints, parameters, and authentication to help you automate voice-over production efficiently.
Documentation
Our documentation covers the full API specification, including request shapes and response formats. You can explore the /docs/ page for clear examples of how to send text and receive audio files for your dubbing projects.
- API reference
- Code snippets
- Endpoint details
Examples
See ready-to-use code samples in various languages. Our /docs/ section includes cURL commands and SDK examples to help you test the API quickly.
- cURL examples
- Python SDK
- JavaScript SDK
Support
For questions or issues, contact us via email. We do not offer phone support or live chat, but our team is ready to assist you with technical queries.
- Email support
- No live chat
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.
