Frequently asked questions
What is text to audio AI?
It is a cloud API that converts text into audio files. You send text via a POST request and receive an audio file in formats like mp3 or wav. The endpoint matches the OpenAI speech endpoint shape, allowing you to use standard SDKs by simply changing the base URL to our API domain.
How does AI voice generation work for training?
Send text to our POST endpoint with your API key. We process the request and return an audio file. Text is split into 4096-character chunks if needed. Requests are processed sequentially, so longer texts may queue. The audio is generated using licensed recordings from LibriVox, ensuring consistent quality for your training materials.
Can I use AI voices for employee onboarding?
Yes. Our voices like Sage (calm, instructional) or Felix (energetic) are ideal for onboarding scripts. You can generate audio for manuals and e-learning content at scale. The API integrates into your existing workflows, allowing HR teams to produce consistent voiceovers without managing complex infrastructure or hiring external voice actors.
Is the audio quality suitable for corporate training?
The audio is generated from licensed recordings, ensuring professional clarity. Voices are distinct presets like Aria or Atlas, designed for clear narration. Each file includes an inaudible watermark marking it as synthetic. The quality is optimized for clarity and comprehension in training environments, supporting 23 languages.
How many languages are supported?
We support 23 languages, including English, Spanish, German, French, Japanese, and Chinese. Any voice can speak any supported language. If the language parameter is omitted, the system guesses from the script, defaulting to English for Latin text. Vietnamese, Indonesian, and Thai are not currently supported.
Can I control the speed of the AI voice?
Yes, you can adjust the speed between 0.25 and 4.0. This allows you to slow down complex training instructions or speed up routine updates. The speed parameter is part of the standard request body, making it easy to tune audio pacing for different types of corporate content.
What happens if a request fails?
If synthesis fails, the charge is returned in full. You are not billed for failed requests. This ensures you only pay for successfully generated audio files. You can retry requests as needed without worrying about extra costs for errors during the generation process.
Do I need a developer to integrate the API?
Our endpoint matches the OpenAI speech endpoint shape. This means you can use the official OpenAI SDKs by changing the base URL. Standard HTTP requests work too. While developers handle integration, the familiar structure reduces the learning curve for teams already familiar with common API patterns.
How does the character limit work?
Each request accepts up to 4096 characters. For longer texts, split them into parts and join the resulting audio files. Spaces and punctuation count as characters. This limit ensures efficient processing. You can manage large training manuals by breaking them into logical segments before sending them to the API.
Are the voices customizable?
Voices are fixed presets: Aria, Nova, Luna, Sage, Felix, Milo, Atlas, and Orion. You cannot clone custom voices or upload your own. You can adjust speed and select from eight distinct tones. Each voice is based on licensed recordings, providing consistent character without the need for custom design or cloning.
Is there a limit on concurrent requests?
Requests are processed one after another, so under load, a request may wait in a queue. You have a rate limit of 60 requests per minute per key. This sequential processing ensures stability. For large-scale training content, plan your generation batches to avoid delays during peak usage.
How do I pay for the service?
Pay $0.01 per 1,000 characters. Top-ups are via USDT on the TRON network (TRC-20), minimum $5. Balance does not expire. No credit cards or subscriptions are used. Start with a $0.50 free trial if you sign in with Google. Failed requests are free.
Can I use this for compliance training videos?
Yes, the API is suitable for compliance content. Use voices like Atlas (neutral, news) for clear delivery. Text is not used for training models, ensuring privacy. Generate audio for manuals and scripts at scale. The consistent voice helps maintain a professional tone across all compliance materials.
Is my text data used for training?
No. Input text is used only to produce the audio and is not used for training. Generated audio is returned in the response and not stored. Only request metadata like time and status is kept for billing purposes. Your training content remains private and secure.
First request
curl
curl https://api.playk1k.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 HR teams choose API-driven voice generation
HR and L&D teams need to turn training manuals, onboarding scripts, and e-learning content into clear audio at scale. An API-driven approach lets you generate voiceovers directly within your existing workflows, removing the need for complex infrastructure or large voice acting teams.
By integrating a reliable text-to-speech API, your team can produce consistent, high-quality narration for hundreds of modules simultaneously. This method ensures every piece of training content sounds professional and uniform, regardless of volume.
Scale your content library
With support for 23 languages and 8 distinct voice presets, you can expand your training library globally without hiring native speakers. Each request generates up to 4096 characters of audio, allowing you to batch-process large documents efficiently.
Audio files are returned after a short generation delay (roughly real-time speed), enabling you to build a vast library of training assets that can be deployed across your organization in minutes rather than weeks.
- Support for 23 languages including English, Spanish, German, and Japanese
- 8 professional voice presets for diverse narration styles
- Process large volumes of text without infrastructure limits
Consistent brand voice
Every generated file uses the same voice model, ensuring that your onboarding, compliance, and leadership modules sound like they belong to the same company. You can select specific voices like "sage" for instructional content or "nova" for upbeat updates to match your tone.
Because the voices are based on licensed recordings from LibriVox, the audio quality remains consistent across all generated files, maintaining a professional standard for your employees.
- Uniform audio quality across all training materials
- Choose from 8 specific voice personalities
- No variation between different narrators or recording sessions
Rapid content updates
When policies change, you don’t need to schedule studio time. Simply send the updated text to the API and receive the new audio file within seconds. This allows HR teams to keep training materials current without delay.
Failed requests are free, so you can regenerate audio for revised text without worrying about extra costs. This reliability makes it easy to iterate on content quickly.
- Generate new audio files in seconds
- Update training content as policies change
- No studio booking required for revisions
8 professional voices for diverse training needs
Our eight voices are designed for corporate training, ensuring clear and engaging audio for diverse learning materials. Each voice offers a distinct tone to match your specific content needs, from gentle onboarding guides to authoritative compliance instructions.
Warm narrators for onboarding
Use Aria for a warm, clear narrator tone or Luna for soft storytelling. These voices create a welcoming atmosphere for new hire introductions and company history modules.
- Aria: Warm, clear narrator
- Luna: Soft, storytelling
Clear instructional voices for compliance
Sage provides a calm, instructional tone ideal for regulatory guidelines. For news-style updates, Atlas offers a neutral delivery that keeps listeners focused on factual content.
- Sage: Calm, instructional
- Atlas: Neutral, news
Engaging tones for leadership modules
- Felix: Energetic, promo
- Milo: Friendly, conversational
- Orion: Deep, documentary

Support for global teams and languages
CoreVoice supports 23 languages, including English, Spanish, French, German, Japanese, and Arabic, allowing HR teams to produce consistent training audio for diverse global workforces. Each voice can speak any of these languages, ensuring uniform tone and quality regardless of region.
Automatic language detection simplifies workflow by inferring the language from the text script, defaulting to English for Latin-script input. For non-English Latin text, explicitly setting the language parameter ensures accurate synthesis.
23 languages supported
Our API supports 23 languages, covering major markets like English, Spanish, French, German, Japanese, Korean, and Arabic. This broad coverage allows you to localize onboarding materials and compliance training without managing multiple voice providers.
- Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish
Automatic language detection
The API automatically detects the language from the input script, defaulting to English for Latin-script text. This reduces configuration overhead for your L&D team, though explicitly setting the language parameter is recommended for non-English Latin texts to ensure precision.
Consistent voice profiles
Each of the eight voices maintains its distinct character across all supported languages. Whether using Aria for warm onboarding narratives or Felix for energetic promo-style training modules, the tonal consistency remains intact, reinforcing your corporate brand identity globally.
Simple integration for L&D workflows
Integrate CoreVoice directly into your L&D pipelines without building custom infrastructure. Our API mirrors the structure of popular speech endpoints, allowing you to leverage existing code and familiar libraries with minimal adjustment.
Simply update your base URL to https://api.playk1k.com/v1 and use your API key. This compatibility means your development team can generate training audio using the tools they already know, reducing onboarding time and technical debt.
OpenAI SDK compatible
Use the official OpenAI SDKs with a simple base_url change. Send standard JSON requests containing your text, voice selection, and speed settings, then receive audio files instantly. This approach works seamlessly with Python, Node.js, and other environments your team already uses.
- Compatible with standard OpenAI client libraries
- Uses familiar JSON request structure
- Works with existing HTTP clients
RESTful API design
Access voices and generate speech via straightforward HTTP endpoints. POST to /v1/audio/speech for generation and GET /v1/voices to check available options. The API returns standard HTTP status codes, making error handling predictable and easy to automate in your CI/CD pipelines.
- Predictable HTTP status codes
- Simple POST/GET endpoints
- Easy automation support
Standard audio formats
Receive audio in MP3, WAV, Opus, FLAC, or PCM formats. Choose the compression level that best fits your LMS bandwidth requirements or archival needs. Each file is ready to play immediately without additional conversion steps, ensuring smooth playback across all major training platforms.
- MP3, WAV, Opus, FLAC, PCM
- No additional conversion needed
- Ready for immediate playback
Transparent pricing for large teams
CoreVoice charges $0.01 per 1,000 characters, ensuring costs scale directly with your training content volume. Failed synthesis requests are free, so you only pay for successful audio generation.
With no monthly minimums or subscription locks, you can adjust usage based on your team’s training cycles without wasting budget on unused capacity.
Pay per character usage
Each character, including spaces and punctuation, counts toward the total. This predictable model helps finance teams forecast expenses accurately for large-scale e-learning deployments.
- $0.01 per 1,000 characters
- Failed requests are free
- Spaces and punctuation count
No monthly minimums
There are no seats, no required minimum spend, and no monthly plans. You top up via USDT on the TRON network, starting at just $5, with no expiration on your balance.
This approach eliminates the risk of paying for idle licenses when training programs are in between updates.
- No monthly plans
- No seats or minimums
- USDT (TRC-20) top-up
Free trial for testing
Sign in with Google to receive $0.50 in trial credit, valid for 7 days. This provides approximately 50,000 characters, allowing your L&D team to test voice quality and integration workflows before committing.
Email-based accounts do not include this trial credit, so use Google sign-in to get started quickly.
- $0.50 free credit
- 50,000 characters
- 7-day validity
- Google sign-in required
How it works: from text to audio
Convert your training materials into audio with a simple POST request. Send text up to 4,096 characters to the speech endpoint, selecting your preferred model and voice. The API handles the synthesis, ensuring high-quality output without manual encoding steps.
Send text via API
Use standard JSON to send your content. Specify the voice, language, and format. If your manual exceeds 4,096 characters, split the text into parts and join the resulting audio files seamlessly.
- POST to /v1/audio/speech
- Supports text up to 4,096 characters
- Optional language code for accuracy
Receive audio files
The API returns the audio file directly in the response. Choose from MP3, WAV, OPUS, FLAC, or PCM formats. Failed requests incur no charge, ensuring you only pay for successful generations.
- Multiple audio formats available
- No charge for failed requests
- Inaudible synthetic watermark included
Integrate with LMS
Download the generated audio files and upload them to your Learning Management System. Use the consistent voice across modules to maintain a professional tone. Automate this process using the API for rapid content updates.
- Direct file download from response
- Consistent voice across modules
- Automate via API integration
Technical reliability for enterprise use
Our infrastructure processes requests sequentially, ensuring stable output even during high-volume training deployments. You can generate roughly 25 characters of text per second, with typical processing times scaling linearly for longer modules. Under heavy load, requests may enter a brief queue, but delivery remains consistent.
Rate limits for stability
Each API key supports 60 requests per minute, providing a predictable ceiling for batch processing. New accounts start with a lower limit of 10 requests per minute, which increases automatically after your first top-up.
- Standard limit: 60 requests/minute
- Initial limit: 10 requests/minute
- Auto-scaling after first top-up
Error handling
Failed synthesis requests are fully refunded, ensuring you never pay for incomplete audio. If a request fails, the charge is returned in full, allowing you to retry without financial penalty.
Billing transparency
Billing is based on character count, including spaces and punctuation, at $10 per 1 million characters. You can monitor usage, balance, and character consumption directly in your dashboard.
Data privacy and security
Your input text is used solely for audio synthesis and is never used to train our models. This ensures that your specific training manuals and proprietary content remain yours, keeping your intellectual property secure within the generation pipeline.
No training on your data
We do not use your input text for model improvement. Once the audio is generated and returned, the text is not retained for training purposes. Your training content remains distinct from our general dataset.
No persistent audio storage
Generated audio files are returned directly in the API response and are not stored on our servers after delivery. This minimizes data retention risks, ensuring your audio assets reside only in your own systems or LMS.
- Audio is returned in the response and not stored
- Input text is not used for training models
- Minimal metadata kept for billing purposes only
Secure key management
Each account holds one active API key for straightforward security. You can issue a new key at any time, instantly invalidating the old one. This simple model prevents unauthorized usage without complex permission hierarchies.

Get started with your first voice
Start your corporate training audio production with a risk-free trial. Create an account using your Google credentials or email to receive $0.50 in credit, valid for 7 days and sufficient for approximately 50,000 characters of generated audio.
Once registered, navigate to the dashboard to generate your unique API key. This key authenticates your requests to the API endpoint, allowing your L&D tools to seamlessly convert training manuals, onboarding scripts, and e-learning content into professional audio files.
Sign up for free trial
Create an account using Google or email/password. You receive $0.50 credit (approx. 50,000 characters) valid for 7 days.
Accounts with email/password start without trial credit. Google sign-in grants immediate access to testing credits.
- Google or email registration
- $0.50 trial credit
- 7-day validity
Generate your API key
Access your unique API key from the dashboard. This key authenticates your requests to the API endpoint.
You can issue a new key at any time; the old one stops immediately. Use it with OpenAI SDKs or standard HTTP.
Try the demo
Use the provided curl example or OpenAI SDK to send text and receive audio. Test voices like Aria or Felix with your content.
Verify output formats (mp3, wav, opus) and ensure integration with your LMS or workflow tools.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Our collective growth depends on embracing new tools, sharing insights, and maintaining open communication across every department and role today.
23 languages
Any voice speaks any of these languages.
