Frequently asked questions
Is this an official Azure text to speech service?
No. We are an independent storefront for the Compliance Audio API, not affiliated with Microsoft, Azure, or any other brand. Our endpoint is https://api.playk138.com/v1/audio/speech. While our request shape matches OpenAI’s, we are a separate service with our own keys, pricing, and infrastructure. You can use standard HTTP clients or the OpenAI SDK by adjusting the base URL.
How does the pricing compare to Azure Cognitive Services?
We charge $10 per 1 million characters, or $0.01 per 1,000 characters. This is a simple, character-based model with no monthly plans or seats. Failed requests are free. Unlike enterprise tools, we do not require minimum spends or complex tiered pricing structures, making it easier to predict costs for legal document generation. We accept USDT (TRC-20) only.
Can I use this API for generating audio from contracts?
Yes. The API accepts up to 4,096 characters per request, which fits many standard contract clauses. For longer documents, split the text into parts and join the resulting audio files. The voices are clear and neutral, making them suitable for reading legal texts, contracts, and compliance documents efficiently.
Do the voices sound natural for legal reading?
The voices are based on licensed recordings from LibriTTS-R and LibriVox public-domain audiobooks. Options like Atlas (neutral, news) or Sage (calm, instructional) are well-suited for legal reading. Each generated file includes an inaudible watermark marking it as synthetic speech.
What happens if my API request fails?
If synthesis fails, the charge is returned in full. You are only billed for successful audio generations. This ensures you do not pay for wasted characters or processing errors. You can retry requests without penalty, making the API reliable for batch processing of legal documents and compliance audio.
Are the voices cloned from real people?
No. The eight voices (Aria, Nova, Luna, Sage, Felix, Milo, Atlas, Orion) are based on licensed recordings from LibriTTS-R and LibriVox public-domain audiobook readings. They are not cloned from specific real individuals for custom use. We do not offer custom voice cloning or uploading your own voice for this service.
Can I upload my own legal voice?
No. We do not offer custom voice cloning or the ability to upload your own voice. You must choose from the eight available presets: Aria, Nova, Luna, Sage, Felix, Milo, Atlas, or Orion. Each voice can speak any of the 23 supported languages, including English, German, and French.
Do you store the legal documents I send?
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. We keep request metadata (time, characters, status) for billing purposes only. Your specific legal content remains transient and is not retained after generation.
Is there a free trial available?
Yes. Signing in with Google provides $0.50 of credit, valid for 7 days, which covers about 50,000 characters. Accounts created with email and password start without trial credit. You can top up with USDT (TRC-20) on the TRON network, with a minimum of $5, credited automatically within minutes.
How many characters can I send per request?
You can send up to 4,096 characters per request. For longer legal documents, split the text into chunks of this size and join the resulting audio files. Spaces and punctuation count as characters. This limit applies to both the tts-1 and tts-1-hd models, which share the same engine and pricing.
Which languages are supported for legal texts?
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. If the language is not set, it is guessed from the script.
Can I use the audio files for commercial projects?
Yes. The API is designed for reliable audio generation for startups and legal documents. The voices are based on licensed recordings (LibriTTS-R, CC BY 4.0), and the generated audio is yours to use. There are no restrictions on commercial use, and the audio files are returned directly in the response for immediate integration into your projects.
Do you offer SLA for uptime?
No. We do not offer an SLA, uptime guarantee, or enterprise contract. Requests are processed one after another, so under load, a request may wait in a queue. Typical speed is roughly real time, but short phrases may take a few seconds. Support is by email only.
How do I switch from Azure to this API?
Update your base URL to https://api.playk138.com/v1 and use the same request shape as OpenAI’s speech endpoint. Replace your Azure key with our Bearer token (tts-...). Use the /v1/voices endpoint to select from our eight presets. No custom SDK is needed; standard HTTP clients or the OpenAI SDK work directly. We do not have an official SDK of our own.
First request
curl
curl https://api.playk138.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 Legal Startups Are Switching
Legal startups often face unpredictable billing spikes when switching from enterprise tools to audio generation. Compliance Audio offers a transparent, character-based pricing model that scales with your actual usage, ensuring you only pay for what you generate.
This approach eliminates the need for complex tiered plans or minimum spend commitments. You can budget accurately for legal document narrations without worrying about hidden fees or sudden rate changes during high-volume periods. Requests are processed sequentially, so under load a request may wait in a queue.
Predictable Cost Model
Our pricing is straightforward: $0.01 per 1,000 characters. This clarity helps you forecast costs for annual report readings or compliance updates without guessing.
Failed requests are free, so you never pay for errors. This protects your budget when testing different voice styles or processing large volumes of text.
- $0.01 per 1,000 characters
- No minimum spend
- Free failed requests
No Enterprise Lock-in
We do not tie you into long-term contracts or expensive seat licenses. Our API is designed for startups that need flexibility and direct access to the technology.
You keep full control over your API keys and can switch providers easily if your needs change, without losing your existing data or workflows.
- No contracts
- No seat limits
- Direct API access
Simple Integration
Our API follows the same request shape as the OpenAI speech endpoint, making it easy to integrate into your existing stack using standard HTTP requests or the OpenAI SDK.
You can start generating audio for legal documents in minutes by changing your base URL and using your API key, reducing development time significantly.
- OpenAI SDK compatible
- Standard JSON body
- Easy setup
8 Professional Voices for Legal Narratives
Our eight voice presets are engineered for clarity, ensuring legal text is delivered without ambiguity. Each voice is based on licensed recordings, providing a professional baseline for your audio workflows.
We group these voices by their primary utility, from clear narration to documentary depth, so you can match the tone to the specific document type without complex configuration.
Clear Narration Voices
Aria, Luna, and Sage offer distinct tones for reading contracts or summaries. Aria provides a warm, clear narrator style ideal for standard terms. Luna brings a soft, storytelling quality to longer clauses. Sage delivers a calm, instructional tone perfect for user guides or compliance manuals.
- Aria: Warm, clear narrator
- Luna: Soft, storytelling
- Sage: Calm, instructional
Conversational Tone
Nova, Felix, Milo, and Orion support more dynamic content. Nova is bright and upbeat for modern tech startups. Felix adds energy for promotional legal updates. Milo offers a friendly, conversational style for client-facing updates.
Orion provides a deep, documentary-style voice for serious regulatory news or annual reports, ensuring the gravity of the content is preserved.
- Nova: Bright, upbeat
- Felix: Energetic, promo
- Milo: Friendly, conversational
- Orion: Deep, documentary
Documentary Style
Atlas and Orion serve neutral to deep roles. Atlas is neutral and news-ready, ideal for unbiased legal briefs. Orion’s deep tone suits documentary-style recitals.
All voices support 23 languages, allowing you to maintain consistent audio quality across global markets without hiring multiple voice actors. Any voice can speak any of these languages.
image]:
Support for Global Legal Markets
Legal teams operate across borders, and your audio needs to reflect that scope. Compliance Audio supports 23 languages, covering major markets from Arabic to Swahili, ensuring your narrated documents reach the right audience without needing separate voice providers.
23 Supported Languages
We cover 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, giving you flexibility for multinational legal filings.
Latin Script Default
If you don’t specify a language, the system guesses based on the script. Latin text defaults to English, which works for most common use cases but might misinterpret mixed-language documents.
Explicit Language Codes
For precision, always set the language field. This prevents errors when processing non-English Latin-script text or ensuring the correct pronunciation for specific regional legal terms in supported languages.
How It Works: Text to Audio
To generate audio, send a POST request to the /v1/audio/speech endpoint with your text and voice selection. The API returns an audio file directly, making it easy to integrate into your legal document workflows without complex setup.
JSON API Body
Send your text in a standard JSON body with fields like model, input, voice, and response_format. The API accepts up to 4096 characters per request, so longer documents need splitting and joining.
OpenAI SDK Compatible
You can use the official OpenAI SDKs by simply changing the base_url to our API endpoint. This allows you to leverage existing code and libraries, reducing the learning curve for your development team.
Fast Synthesis
Typical speed is roughly real-time, generating about 25 characters of text per second of audio. Short phrases take a few seconds to synthesize. Requests are processed sequentially, so under load, a request may wait in a queue.
Pricing Transparency
Our pricing is straightforward: $10 per 1 million characters, or $0.01 per 1,000 characters. Spaces and punctuation count toward the total. This character-based model scales linearly, making budget forecasting for large legal document batches predictable and simple.
Per Character Rate
You pay only for what you generate. At $0.01 per 1,000 characters, even lengthy contracts remain cost-effective. The rate applies uniformly across all models (tts-1 and tts-1-hd) and voices, so you can choose quality without worrying about tiered pricing structures.
Free Failed Requests
If synthesis fails for any reason, the charge is returned in full. You are not billed for errors, ensuring you only pay for successfully generated audio files.
No Hidden Fees
We do not charge for API calls, storage, or additional features. There are no VAT invoices and no hidden costs for extra languages or formats. Top-ups via USTRC-20 are credited automatically, with no minimum spend or monthly subscriptions to manage.

Data Privacy for Sensitive Docs
For legal startups, the confidentiality of client data is non-negotiable. Our API is designed to respect the sensitivity of your documents by ensuring that the text you submit is used solely for the immediate generation of audio. We do not use your input data to train our models, meaning your specific legal narratives remain yours.
No Training on Input
Unlike many enterprise solutions, we do not retain your input text for machine learning purposes. Once the audio is generated, the original text is not stored in our training datasets, ensuring that proprietary case details or client information remain isolated from our model development.
No Audio Storage
Generated audio files are returned directly in the response and are not stored on our servers after delivery. This means you maintain full control over where your audio files are saved, whether in your own cloud storage or internal servers, without us holding copies of your final deliverables.
Metadata Only
We keep only the essential metadata required for billing and debugging, such as the timestamp, character count, and request status. This minimal data retention ensures that your usage history is accurate without exposing the content of your legal documents to our storage infrastructure.
Technical Limits You Should Know
Legal documents often exceed character limits, so you will need to split larger texts into 4096-character chunks before sending them to the API. Once generated, these audio segments can be joined to form the complete narration.
Keep in mind that requests are processed sequentially. If your startup sends many requests at once, they may wait in a queue rather than processing simultaneously.
4096 Character Limit
Each API call accepts a maximum of 4096 characters. This includes spaces and punctuation, so a dense legal clause might hit the limit faster than expected. You must split longer documents into smaller parts and join the resulting audio files.
Rate Limits
Your API key allows 60 requests per minute. New accounts start with a lower limit of 10 requests per minute until the first top-up. Exceeding this limit will result in errors until the window resets.
Queue Processing
Requests are processed one after another. Under heavy load, your synthesis jobs may wait in a queue, increasing total generation time. For most legal use cases, this sequential processing ensures stable and predictable output.
Get Started in Minutes
Start by creating an account on the /get-api-key/ page using your Google sign-in. This grants you an immediate free trial of $0.50, enough to generate roughly 50,000 characters of audio. Accounts created with email and password start without trial credit. No credit card is required to begin testing your legal document workflows.
Sign Up with Google
Click the sign-up button on our registration page and authorize access with your Google account. Your account is created instantly, and the trial credit is applied automatically. This method is the fastest way to start testing without entering payment details or waiting for email verification.
Get API Key
Once logged in, navigate to the dashboard to view your unique API key. This key authenticates your requests to the endpoint. You can issue a new key at any time, which instantly invalidates the old one, ensuring your credentials remain secure as your startup scales.
- Keys are displayed clearly in the dashboard.
- Revoking a key stops it immediately.
- One active key per account.
First Request
With your key ready, send a POST request to https://api.playk138.com/v1/audio/speech. Use the OpenAI SDK by changing the base URL, or use plain HTTP. Send your legal text in the input field, select a voice like 'sage' for calm narration, and receive the audio file directly in the response.
- Endpoint: POST https://api.playk138.com/v1/audio/speech
- Auth: Bearer tts-... key
- Format: JSON body
Frequently Asked Questions
We are a standalone API, not affiliated with Microsoft or Azure. While Azure offers SSML and premium voice models, we provide a simpler, character-based pricing model with 8 clear voices and 23 languages, designed for startups that prefer predictable costs over enterprise lock-in.
Our API mirrors the OpenAI speech endpoint structure, making migration straightforward. Unlike Azure’s complex tiered subscriptions, we charge only $0.01 per 1,000 characters with no minimum spend, and failed requests are free.
Azure vs Our API
Azure provides SSML, extensive voice customization, and enterprise SLAs. We offer a lightweight alternative: no SSML, no custom voice cloning, and no SLA, but with significantly lower costs and faster integration for standard legal narration needs.
Voice Cloning
We do not offer custom voice cloning or uploading your own voice samples. Our eight presets (Aria, Nova, Luna, Sage, Felix, Milo, Atlas, Orion) are based on licensed LibriTTS-R recordings and public-domain audiobook readings, providing consistent, high-quality narration without the need for training data.
Legal Compliance
Input text is used solely for generation and is not used for training models. Generated audio is returned in the response and not stored on our servers. Request metadata (time, characters, status) is retained only for billing purposes. We do not provide specific legal certifications like HIPAA or GDPR certificates, but our data handling is transparent and minimal.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Teams must carefully monitor character counts to avoid unexpected costs.
23 languages
Any voice speaks any of these languages.
Sources and further reading
- Wikipedia
- Creative Commons Attribution 4.0
- curl documentation
- ISO 639 language codes
- IETF RFC 7845: Ogg Opus
- TRON developer hub: TRC-20
- Tether (USDT)
- W3C WAI: making audio and video accessible
- OpenAI API reference: create speech
- LibriVox public-domain audiobooks
- openai-node on GitHub
- LibriTTS-R corpus (OpenSLR 141)
