Frequently asked questions
Which voice is best for legal documents?
Q: Which voice is suitable for legal documents? A: Atlas (neutral, news-style) and Sage (calm, instructional) are well-suited for dense legal text. Aria (warm narrator) and Orion (deep, documentary) are also viable options. All eight voices support any of the 23 languages, allowing you to select the tone that best fits your specific compliance or client communication needs.
How do I calculate the cost of a contract?
Pricing is $10 per 1 million characters, or $0.01 per 1,000 characters. Spaces and punctuation count as characters. For example, a 50,000-character contract costs $0.50. Failed requests are free, so you only pay for successful synthesis. You can track your exact usage and remaining balance directly in your account dashboard after signing up.
Can I use the API for client notifications?
Yes. You can send text via POST to the speech endpoint and receive an audio file. This works for notifications, provided the text is under 4,096 characters per request. You can adjust the speed between 0.25 and 4.0 to ensure clarity. The audio files are standard formats like MP3 or WAV, easily embedded in emails or apps.
Do I need to split long contracts manually?
You must split texts longer than 4,096 characters into parts yourself. Each request handles up to that limit. After receiving the audio files for each part, you join them to create the final document. This ensures consistent billing and processing, as each segment is treated as an independent synthesis request with its own character count.
Is the audio quality suitable for legal use?
The API uses models tts-1 or tts-1-hd for clear speech. Voices are based on licensed recordings from LibriTTS-R and LibriVox. While synthetic speech includes an inaudible watermark, the audio is designed for clarity and comprehension. It is suitable for listening to contracts, terms, and conditions, though it does not include emotional nuance tags.
Can I store the generated audio files?
Yes. The API returns the audio file in the response but does not store it on our servers. You are responsible for storing the generated files in your own infrastructure. Input text is used only for generation and not for training. Request metadata is kept for billing purposes, but the audio content itself is yours to keep.
Do you support multiple languages in one contract?
Each request supports one language via the optional language field. If you do not set it, the system guesses the language, defaulting to English for Latin script. For contracts mixing languages, you should split the text by language and send separate requests. Supported languages include English, German, French, Spanish, and others, but not Vietnamese, Thai, Ukrainian, Romanian, Czech, Bengali and all others not in the list.
How do I handle errors during generation?
If synthesis fails, the charge is returned in full. You can retry the request or split the text if it exceeds limits. Check the response status for details. Since requests are processed sequentially, high load may cause delays. Ensure your text is under 4,096 characters and your API key is valid before sending the POST request.
Is there a limit on the number of contracts?
There is no limit on the number of contracts or documents you can process. Limits apply to rate usage: 60 requests per minute per key. You can issue a new key if needed. There are no monthly plans, seats, or minimum spends. You pay only for the characters you successfully generate, with no expiration on your balance.
Can I clone a specific lawyer's voice?
No, custom voice cloning is not offered. You choose from eight preset voices like Atlas, Sage, or Aria. These are based on licensed recordings, not specific individuals. You cannot upload your own voice or design a custom one. The voices are synthetic and include an inaudible watermark to mark them as generated by the API.
Do you provide an SLA for production?
No, there is no Service Level Agreement (SLA) or uptime guarantee. Requests are processed one after another, so under load, a request may wait in a queue. Typical speed is roughly real-time for long texts. Support is available by email only; there is no phone support, live chat, or enterprise contract for guaranteed performance.
How do I integrate with my existing app?
Send a POST request to https://api.playvt443.com/v1/audio/speech with a JSON body containing model, input, voice, and response_format. Use your API key in the Authorization header. You can also use the official OpenAI SDK by changing the base_url to https://api.playvt443.com/v1. This allows you to leverage existing code structures with minimal changes.
Understanding Legal Audio Needs
Legal teams require audio outputs that maintain the precise integrity of dense source text. Legal documents demand accurate character counting because billing is strictly per character, including spaces and punctuation. This transparency ensures that compliance-ready documentation can be audited without hidden costs or ambiguity in volume estimates.
Compliance requirements
ClarityVoice returns generated audio in the response and does not store it, reducing data retention risks. Input text is used solely for synthesis and is not used for training, ensuring that sensitive contract details remain within your immediate workflow. Request metadata is kept only for billing accuracy.
- No audio storage after response
- Input text not used for training
- Metadata retained only for billing
Accessibility standards
Audio generation supports 23 languages, covering major legal jurisdictions. Each generated file carries an inaudible watermark marking it as synthetic speech, which can be useful for internal compliance tracking. Teams can select voices like Atlas for neutral news-style reading or Sage for calm, instructional delivery of complex clauses.
- Supports 23 languages for global teams
- Inaudible synthetic watermark for tracking
- Neutral voice options for clarity
User experience goals
Legal professionals often listen to lengthy documents during commute or review. The API generates audio at roughly real-time speed, producing about 25 characters of text per second. This natural pace helps reviewers absorb dense regulatory updates without the fatigue associated with prolonged visual reading.
- Real-time generation speed (~25 chars/sec)
- Reduces visual fatigue for reviewers
- Ideal for background listening during review
Comparing Legal Voices
Selecting the correct voice preset ensures legal content is both legally compliant and accessible. The choice between neutral and warm tones impacts how dense contractual language is perceived by end-users.
We evaluate four key voice presets to match specific legal contexts. This guide helps product teams align audio output with document type and audience expectations.
Neutral vs. Warm tone
Neutral voices like Atlas provide clear, unbiased narration ideal for regulatory updates or news-style compliance documents. Warm voices like Aria or Luna add a softer, more engaging quality suitable for client-facing notifications or explanatory legal guides.
Atlas delivers a neutral, news-style delivery that minimizes emotional bias. Aria offers a warm, clear narrator tone that makes dense text feel more approachable without sacrificing clarity.
- Atlas: Neutral, news-style delivery
- Aria: Warm, clear narrator tone
Instructional vs. Narrative
Instructional voices like Sage are optimized for clarity and pacing, making them ideal for step-by-step compliance instructions or user manuals. Narrative voices like Luna or Nova suit longer, story-like legal summaries or case study explanations.
Sage provides a calm, instructional tone perfect for procedural documents. Luna offers a soft, storytelling voice that enhances engagement for narrative-driven legal content.
- Sage: Calm, instructional tone
- Luna: Soft, storytelling voice
Choosing the right preset
Match the voice to the document’s purpose. Use Atlas for unbiased updates, Sage for instructions, and Luna for engaging summaries. Every voice can speak any of the 23 languages.
Test samples with your specific legal text to ensure clarity. Voices are based on licensed recordings from LibriTTS-R and LibriVox.
- Atlas: Unbiased updates
- Sage: Instructions
- Luna: Engaging summaries
Pricing for Legal Volumes
Legal teams often generate substantial audio volumes from dense documentation. Our billing is strictly character-accurate, costing $10 per 1 million characters, or $0.01 per 1,000 characters. Spaces and punctuation count toward this total, ensuring predictable costs for compliance-ready documentation.
Cost per 1,000 characters
Pricing scales linearly with text length. For example, a 10,000-character contract section costs $0.10. Failed requests are free, so you only pay for successful synthesis. This structure supports high-volume legal workflows without hidden fees.
- $0.01 per 1,000 characters
- Spaces and punctuation count
- Failed requests are free
Estimating large contracts
Estimate volume by character count rather than duration. Long texts process at roughly real-time speed, taking about 25 characters of text per second of generation. Short phrases may take a few seconds due to queue processing.
- Linear pricing model
- Real-time generation speed
- No minimum spend
Budgeting with USDT
Top up using USDT on the TRON network (TRC-20) only, with a $5 minimum. Credits are automatic within minutes. There are no subscriptions, invoices, or VAT. Balance does not expire, allowing flexible budgeting for legal departments.
- USDT on TRON (TRC-20) only
- $5 minimum top-up
- No monthly plans
Technical Integration Steps
Integration begins by authenticating your requests with a Bearer token. Navigate to /get-api-key/ to generate your key, which is tied to your account balance and rate limits. Use this key in the Authorization header for every call to the API base URL.
API key setup
Generate your key via the /get-api-key/ dashboard. Each account holds one active key; issuing a new one immediately invalidates the previous one. Include Authorization: Bearer tts-... in your request headers.
SDK configuration
The endpoint mirrors the OpenAI speech structure. Point your OpenAI SDK base_url to https://api.playvt443.com/v1 and use your ClarityVoice key. This allows you to use the official OpenAI SDKs without custom client code.
Error handling
Monitor HTTP status codes. Failed synthesis returns a 4xx/5xx error and refunds the character cost. Ensure your code handles these gracefully to maintain accurate billing records without losing credit for invalid inputs.
Handling Long Texts
Legal documents often exceed the 4,096-character limit per request. To handle contracts or long regulatory updates, you must split the text into smaller chunks before sending them to the API. This ensures each segment is processed correctly without truncation or errors.
Once synthesized, these individual audio files need to be concatenated into a single continuous file for your end users. Proper sequencing is critical to maintain the logical flow of the legal document during playback.
Splitting strategies
Divide your input text at logical boundaries, such as sentence endings or clause breaks, ensuring no chunk exceeds 4,096 characters. This prevents partial word cuts that could confuse legal terminology. Keep track of the original order to reconstruct the sequence accurately later.
The API accepts plain text in the input field. You can implement this splitting logic in your backend code before making the POST requests. This approach keeps the API calls efficient and within character limits.
- Split at sentence or paragraph boundaries
- Ensure each chunk is <= 4,096 characters
- Maintain original text order
Joining audio files
After receiving the audio responses, concatenate the files in the order they were processed. Requests are processed one after another, so maintaining the sequence is straightforward. Tools like FFmpeg or your preferred audio library can merge mp3, wav, or opus files seamlessly.
This creates a single, listenable audio file that represents the entire document. It ensures a smooth listening experience for your users without awkward pauses between segments.
- Merge files in original order
- Use FFmpeg or audio libraries
- Output single continuous file
Queue management
Requests are processed one after another, so under load, a request may wait in a queue. For large documents, this can add latency. Monitor your queue depth to estimate total generation time.
Typical speed is roughly real time (25 characters per second). Plan your infrastructure to handle the queue without timing out. The rate limit is 60 requests per minute per key, so batch your requests accordingly.
- Requests are queued sequentially
- Rate limit: 60 requests/minute
- Monitor queue for latency
Language Support for Legal
Legal teams often handle multilingual contracts or international regulatory documents. ClarityVoice supports 23 languages, ensuring your audio outputs match the source text’s origin. This flexibility is critical for global compliance and client communication.
However, automatic language detection has limits. If the input text uses Latin script but isn’t English, the system defaults to English unless specified. For precise results, always define the language explicitly.
23 supported languages
Our 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.
Every voice preset can speak all 23 languages, allowing consistent tone across diverse documents.
- Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish
Auto-detection limits
If the language field is omitted, the system guesses based on script. Latin script defaults to English, which may cause mispronunciation for other Latin-based languages like French or Spanish.
Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, Bengali and all others not in the list are not supported.
Explicit language codes
For non-English Latin-script text, always set the language parameter to the correct ISO code. This ensures accurate pronunciation and maintains the professional tone required for legal audio.
This practice prevents errors in contract readings or regulatory updates where precise enunciation is vital.
Data Privacy Considerations
ClarityVoice is engineered for legal teams who need strict data boundaries. When you send dense text for synthesis, that input is processed exclusively to generate the audio file. We do not use your contract language, case details, or client names to train our models or improve future outputs. This ensures your proprietary information remains yours.
The generated audio is returned directly in the API response and is not stored on our servers after delivery. This means your synthesized legal documents do not linger in our storage, reducing the attack surface for data breaches. You maintain full control over where that audio is saved and how long it is retained in your own systems.
No training on input
Your text input is used solely for the immediate generation of the audio file. Unlike many consumer-grade TTS services, we do not ingest your legal texts into our training datasets. This means your specific phrasing, terminology, and confidential case details do not influence the voice model's future behavior or output for other users.
- Input text is ephemeral and used only for synthesis.
- No usage of input data for model training.
- Proprietary legal language remains isolated to your requests.
No audio storage
Once the audio file is delivered to your client application via the API response, we do not retain a copy of the generated audio. This minimizes data residency concerns and ensures that your synthesized legal audio does not accumulate in our infrastructure. You are responsible for storing the audio files according to your own compliance requirements.
- Audio is returned in the response and not stored by ClarityVoice.
- No long-term retention of generated audio files.
- Reduces risk of unauthorized access to synthesized content.
Metadata retention
We do keep request metadata for billing and operational purposes. This includes the timestamp of the request, the number of characters processed, and the status of the synthesis (success or failure). This data is necessary to accurately charge your account and troubleshoot any issues with your API usage, but it does not include the actual text content or the resulting audio.
- Timestamps are recorded for each request.
- Character counts are stored for accurate billing.
- Status codes (success/failure) are retained for operational support.

Common Use Cases
Legal teams use TTS to convert dense documentation into clear audio, enabling attorneys to review contracts during commutes or while multitasking. This transforms static text into listenable, compliant documentation without requiring dedicated listening time.
The API handles character-accurate billing, ensuring precise cost tracking for high-volume regulatory updates and client notifications. Teams can process long texts by splitting them into 4096-character parts, joining the audio seamlessly.
Contract review
Attorneys can listen to lengthy terms and conditions while working elsewhere. The neutral tones of voices like Atlas or Sage ensure clarity without distraction, making complex legal jargon easier to digest audibly.
- Listen to contracts during commutes
- Clarify complex legal jargon
- Reduce screen fatigue
Regulatory updates
Keep teams informed with timely audio updates on changing regulations. The API’s reliable speed—roughly real-time—ensures updates are delivered quickly, allowing legal staff to stay compliant without reading every document manually.
- Timely compliance notifications
- Real-time delivery speed
- Multi-language support
Client notifications
Send clear, professional audio messages to clients regarding case updates. Using voices like Felix or Luna, legal teams can enhance client engagement with accessible audio updates.
- Professional client updates
- Personalized audio messages
- Accessible communication
Troubleshooting Guide
Legal teams often process large volumes of text for compliance or client updates. When integration hiccups occur, they typically stem from three specific areas: rate limits, character boundaries, or format mismatches. Understanding these constraints prevents delays in your audio delivery pipeline.
Rate limit issues
You are allowed 60 requests per minute per API key. If you exceed this, requests will queue or fail. Before your first top-up, the limit is stricter at 10 requests per minute.
- Implement exponential backoff in your integration code.
- Monitor your usage dashboard at /get-api-key/ for real-time limits.
Character count errors
Each API request accepts a maximum of 4096 characters. This count includes spaces and punctuation. If your legal document exceeds this, you must split the text into smaller chunks before sending them to the endpoint.
- Validate character length in your preprocessing layer.
- Ensure your split logic preserves paragraph boundaries where possible.
Audio format issues
Specify your desired format (mp3, wav, opus, flac, or pcm) in the response_format field. If omitted, the API defaults to mp3. Ensure your playback system supports the chosen codec to avoid silent errors.
- Test with multiple formats during integration.
- Verify file headers match the requested format.
Best Practices for Legal Teams
Legal teams must ensure text-to-speech outputs accurately reflect the source document’s nuance and structure. By following a few disciplined practices, you can maintain consistency across large volumes of generated audio while keeping costs predictable.
Clean input text
Ensure your source text is free of extraneous formatting characters that might confuse the synthesizer. Since billing is based on character count, removing unnecessary spaces or punctuation reduces costs without affecting comprehension. Keep paragraphs logical to maintain audio flow.
- Remove non-standard symbols before sending to the API.
- Verify paragraph breaks align with natural speech pauses.
- Ensure character count includes all spaces and punctuation.
Monitor usage
Track your character consumption closely, as billing is precise at $0.01 per 1,000 characters. Use the dashboard to monitor your balance and request limits. This helps avoid unexpected charges when processing large legal contracts or regulatory updates.
- Check balance regularly via the dashboard.
- Set alerts for approaching rate limits.
- Review character counts in API responses.
Test with samples
Before processing entire documents, test short excerpts with different voices. This ensures the chosen tone matches your legal context, whether it’s a neutral news style or a warm narrative. Adjust speed and language settings to optimize clarity for your audience.
- Test voices like Atlas for neutrality or Aria for clarity.
- Verify audio quality with short legal excerpts.
- Adjust speed settings for complex terminology.
Getting Started with Trial
ClarityVoice offers a frictionless entry point for legal teams evaluating the API’s fidelity for dense contractual text. By signing in with Google, you instantly access your dedicated dashboard without navigating complex enterprise onboarding flows or waiting for invoice approvals.
Sign in with Google
Navigate to the /get-api-key/ page and select the Google sign-in option. This method creates an account on playvt443.com, linking your identity to a unique API key issued immediately upon authentication.
- Instant account creation without email verification delays
- Secure OAuth 2.0 authentication
- Automatic key generation for immediate testing
Access free credit
Upon successful Google sign-in, your account is credited with $0.50 of trial funds. This balance provides approximately 50,000 characters of synthesis, sufficient for testing a standard non-disclosure agreement or a brief compliance update.
- Credit expires in 7 days
- No credit card required upfront
- Valid for all 8 voice presets
Run first test
Use the provided cURL example on your dashboard to send a POST request to the speech endpoint. Input a short legal clause to verify audio quality, latency, and character billing accuracy before scaling to larger contracts.
Support and Contact
ClarityVoice provides dedicated support via email for all technical and billing inquiries. Unlike traditional support models, we do not offer a phone number or live chat, ensuring that every ticket is documented and traceable for your records. This asynchronous approach is ideal for legal teams managing complex integration workflows without the pressure of immediate response times.
Our email support team is equipped to assist with API key issues, USDT top-up confirmations, and general usage questions. For faster resolution, please include your API key and a description of the issue in your initial message.
Email support
Send your queries to [email protected]. We aim to respond to all tickets within one business day. For urgent billing issues regarding USDT transfers, please include the transaction hash in your email to expedite verification.
Our support team can also assist with troubleshooting rate limits or clarifying character counting discrepancies in your billing history.
- Single point of contact via email
- No phone or live chat available
- All communications are logged for audit purposes
Documentation
The complete API reference is available at /docs/. This guide covers all endpoints, including the speech generation endpoint and the voices endpoint. It includes ready-to-use curl examples and explains the JSON request structure in detail.
You can also find specific integration notes for the OpenAI SDK and details on error codes and retry strategies directly in the documentation.
- Full API reference at /docs/
- Includes curl examples and SDK guides
- Covers error codes and retry logic
Community resources
While we do not maintain a public forum, developers often share tips and integration snippets on technical blogs and GitHub. You can also check the /about/ page for more background on ClarityVoice and our commitment to transparent, character-accurate billing for legal teams.
For self-service, the /docs/ page is the most comprehensive resource for understanding rate limits, voice presets, and supported languages.
- No official public forum
- External developer blogs and GitHub repositories
- Check /about/ for company background
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. This ensures each segment is processed correctly without truncation or errors.
23 languages
Any voice speaks any of these languages.
