Frequently asked questions
What is the difference between TTS-1 and TTS-1-HD models?
Both models use the same engine and incur the same price of $10 per 1 million characters. The distinction lies solely in audio quality, with TTS-1-HD offering higher fidelity. You may select either model via the model field in your request without any change to your billing rate or character limits.
Which male voice is best for IVR navigation?
Atlas is a neutral, news-style voice suitable for standard IVR prompts. Milo offers a friendly, conversational tone. Felix provides an energetic sound for promotional messages. All voices support the 23 supported languages, allowing you to select the preset that fits your system's tone.
How does the character counting work for billing?
Billing is based on a rate of $10 per 1 million characters, or $0.01 per 1,000 characters. Spaces and punctuation count toward the total character count. Each request processes up to 4096 characters. Failed requests incur no charge, and the credit is returned in full if synthesis fails.
Can I use this API for commercial call center recordings?
Yes. The voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain audiobook readings. The API provides synthetic speech suitable for commercial use. Generated audio is not stored, and input text is used only to produce the audio, ensuring clear usage rights for your call center applications.
What happens if my API request fails?
If synthesis fails, the charge is returned in full. You will not be billed for characters processed during a failed request. This applies to any request where the audio file is not successfully generated and returned in the response, ensuring you only pay for successful audio production.
Does the API support automatic language detection?
The API guesses the language from the script if the language field is not set. Latin script text defaults to English. For non-English Latin-script text, you should explicitly set the language parameter. It supports 23 specific languages, including English, Spanish, German, and Japanese, among others.
How long does it take to receive an audio file?
Typical speed is roughly real time, generating about 25 characters of text per second. Short phrases may take a few seconds. Requests are processed sequentially, so under load, a request may wait in a queue. Longer texts are split into parts and joined, which may increase total time.
Is there a limit on concurrent requests?
The rate limit is 60 requests per minute per key, or 10 per minute before the first top-up. Requests are processed one after another, meaning concurrent requests may queue if the limit is exceeded. There is no specified limit on simultaneous connections, but the per-minute request cap applies strictly.
Can I generate audio files larger than 4096 characters?
Yes, but you must split the text into parts of up to 4096 characters per request. You then join the resulting audio files. The API does not handle this splitting automatically; your application must manage the chunking logic and concatenate the returned audio streams or files.
Are the voices based on real human recordings?
Yes. The voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain audiobook readings. Every generated file carries an inaudible watermark marking it as synthetic speech. The voices are presets derived from these sources, not custom-cloned from specific individuals for your account.
Do I need to attribute the voices in my app?
The voices are based on LibriTTS-R (CC BY 4.0) and LibriVox public-domain readings. Attribution requirements depend on the specific license of the source recording and your distribution method.
How do I switch from a free trial to a paid account?
Sign in with Google to receive a $0.50 trial credit valid for 7 days. To add funds, top up with USDT on the TRON network (TRC-20), minimum $5. Balance does not expire. Top-ups are credited automatically within minutes. No subscriptions, invoices, or VAT are involved; you pay as you use.
What audio formats are supported for production?
The API supports mp3, wav, opus, flac, and pcm formats. You specify the desired format using the response_format field in your request. These formats cover most production needs for call centers and IVR systems, ensuring compatibility with various telephony platforms and audio processing pipelines.
Is there an SLA for uptime guarantees?
No. There is no SLA, no uptime guarantee, and no enterprise contract. The service is provided as-is without guaranteed availability. Requests may queue under load, and there is no on-premise install option or data-center choice. Support is available by email only, with no phone or live chat.
First request
curl
curl https://api.playmvpviva.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.mp3Core API Capabilities
Viva Voice Labs provides a robust text-to-speech API designed for call center operators and indie developers building high-volume IVR systems. We deliver consistent, licensable synthetic voices with transparent character-based pricing and stable rate limits.
OpenAI-Compatible Endpoint
Send text and receive an audio file via a standard POST request to our endpoint. The request shape matches the OpenAI speech endpoint, allowing official OpenAI SDKs to work by changing the base URL to https://api.playmvpviva.com/v1.
- Endpoint: POST https://api.playmvpviva.com/v1/audio/speech
- Uses standard JSON body structure
- Compatible with existing OpenAI SDK implementations
Multi-Language Support
Our engine supports 23 languages, including English, Spanish, German, French, Chinese, Japanese, and Arabic. Any voice can speak any of these languages, ensuring consistent audio quality across your global IVR network.
- Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish
Standardized Request Format
Requests require a model selection (tts-1 or tts-1-hd), input text up to 4096 characters, and a voice preset. You can specify the response format (mp3, wav, opus, flac, pcm) and adjust speed between 0.25 and 4.0.
- Models: tts-1, tts-1-hd
- Max input: 4096 characters
- Formats: mp3, wav, opus, flac, pcm
Voice Catalog for Call Centers
Viva Voice Labs provides eight distinct voice presets optimized for professional IVR environments, ensuring consistent audio quality across high-volume call centers. Our catalog includes both male and female options, allowing developers to select the most appropriate tone for their specific application needs without requiring custom voice design.
Male Voice Presets
We offer three male voices designed for distinct communication styles. Felix delivers energetic delivery suitable for promotional messages, while Milo provides a friendly, conversational tone ideal for general inquiries. Orion offers a deep, documentary-style voice for formal announcements.
- Felix: Energetic, promo
- Milo: Friendly, conversational
- Orion: Deep, documentary
Female Voice Presets
Our female voice options include Aria for warm, clear narration and Nova for bright, upbeat interactions. Luna provides a soft, storytelling quality, while Sage delivers calm, instructional audio perfect for step-by-step guidance in automated systems.
- Aria: Warm, clear narrator
- Nova: Bright, upbeat
- Luna: Soft, storytelling
- Sage: Calm, instructional
Neutral and Deep Options
Atlas serves as a neutral voice option, particularly effective for news-style updates or standard informational prompts where a distinct personality is not required. All voices support 23 languages, allowing for flexible localization without additional licensing fees. Any voice can speak any of these languages.
- Atlas: Neutral, news
Technical Infrastructure
Viva Voice Labs uses a transparent character-based pricing model at $10 per million characters, with spaces and punctuation counting toward the total. Failed requests incur no charge, ensuring you pay only for successful synthesis.
Character-Based Pricing
Each request is billed based on the exact character count of the input text. This predictable model eliminates surprise costs, making it easy to forecast expenses for high-volume IVR deployments.
Audio Formats and Speed
Output is delivered in mp3, wav, opus, flac, or pcm formats. You can adjust the playback speed between 0.25x and 4.0x to fit specific IVR timing requirements without re-generating audio.
Rate Limits and Queuing
Standard accounts support 60 requests per minute (10 per minute before the first top-up). Requests are processed sequentially, so high-volume bursts may experience queuing during peak load, but long-text generation remains near real-time.
Use Cases for Indie Developers
Indie developers building IVR systems require consistent, licensable synthetic voices that integrate seamlessly with standard telephony workflows. Our API delivers high-volume audio generation with transparent character-based pricing, allowing you to scale greeting messages and notification systems without complex infrastructure management or unpredictable rate limits.
IVR Greeting Messages
Generate precise welcome sequences using male presets like Atlas for neutral news-style intros or Felix for energetic promo tones. The API accepts up to 4096 characters per request; longer scripts must be split and joined. Audio formats like MP3 or WAV ensure broad compatibility with existing VoIP gateways and PBX systems.
- Consistent tone across all customer touchpoints
- Scalable generation for high call volumes
- Standardized JSON request format for easy integration
On-Hold Audio Content
Create looping on-hold music or informational messages with precise speed control (0.25–4.0x). Use voices like Orion for deep, documentary-style updates or Milo for conversational announcements. Each request generates a distinct audio file that can be concatenated server-side to create extended hold experiences.
- Adjustable speed for optimal hold times
- Support for 23 languages including English and Spanish
- No SSML required for basic text-to-speech
Notification Systems
Trigger automated voice notifications for appointment reminders, shipping updates, or system alerts. The API processes requests sequentially, ensuring reliable delivery for critical communications. Failed synthesis requests incur no charge, reducing risk for high-volume notification pipelines. Integrate via standard HTTP POST calls using any language's official SDK or plain HTTP.
- Charge only for successful synthesis
- Simple curl or SDK integration
- Reliable sequential processing for steady throughput
Licensing and Data Privacy
Viva Voice Labs ensures that all synthetic voices are derived from licensed recordings, providing a clear legal foundation for commercial IVR deployments. Our audio outputs are generated from established sources, allowing developers to integrate voice assets with confidence in their licensing status without navigating complex rights chains.
Regarding data privacy, input text is utilized exclusively for the immediate synthesis of audio files and is not retained for model training. This ensures that your customer data remains isolated to the specific transaction, reducing exposure risks in high-volume call center environments where privacy compliance is critical.
Licensed Voice Recordings
Each voice preset is based on licensed recordings from LibriTTS-R (CC BY 4.0) and public-domain audiobook readings from LibriVox. This foundation provides a stable, licensable asset base for both indie developers and larger call center operations, ensuring that the synthetic voices used in your IVR systems are legally sound for commercial use.
- Based on LibriTTS-R (CC BY 4.0)
- Public-domain source material from LibriVox
- Stable licensing for commercial IVR use
Input Data Usage
Input text is processed solely to generate the requested audio file. We do not use your input text for training our models, ensuring that your specific customer data or script content remains private to your organization. Request metadata, including time and character count, is retained only for accurate billing purposes.
- Text used only for synthesis
- No text used for model training
- Metadata kept for billing only
Synthetic Watermarking
Every generated audio file contains an inaudible watermark that explicitly marks the content as synthetic speech. This watermarking aids in identifying AI-generated audio within your IVR systems or broadcast materials, providing an additional layer of transparency and compliance for your audio assets.

Getting Started
Access the /get-api-key/ page to generate your API credentials. You may sign in using Google for an immediate $0.50 trial credit or create an email-based account. The dashboard displays your active key, current balance, and usage metrics.
API Key Generation
Upon signing in, you receive a single active bearer key. You can issue a new key at any time; the previous key becomes invalid immediately. This ensures secure rotation without downtime for your IVR integration.
- Google sign-in grants $0.50 trial credit
- Email sign-in starts with zero balance
- One active key per account
First Request Demo
Send a POST request to https://api.playmvpviva.com/v1/audio/speech with your bearer token. Use the 'tts-1' model and select a male voice like 'atlas' for news-style IVR greetings. The endpoint returns the audio file in the response body.
- Endpoint: POST /v1/audio/speech
- Auth: Authorization: Bearer tts-...
- Model: tts-1 or tts-1-hd
Supported Protocols
The API uses standard HTTPS with JSON payloads, compatible with the OpenAI SDK by adjusting the base_url. Other endpoints include GET /v1/voices for catalog details and GET /v1/models for available engines. No proprietary SDK is required.
Comparison Standards
Viva Voice Labs offers two model variants, tts-1 and tts-1-hd, which utilize the same underlying engine and pricing structure. The distinction lies solely in the audio fidelity of the generated synthetic speech files.
Unlike streaming protocols that deliver partial audio chunks, this API delivers complete audio files upon request completion. This ensures consistent file integrity for IVR systems that require static audio assets rather than real-time partial streams.
TTS-1 vs TTS-1-HD
Both models process text through identical synthesis engines and share the same character-based pricing of $0.01 per 1,000 characters. Developers may select either model without affecting cost or rate limits.
The primary difference is audio quality. tts-1-hd and tts-1 use the same engine and pricing; the distinction lies solely in audio fidelity.
- Same engine and pricing
- Identical rate limits (60 req/min)
- Select via model parameter
Streaming vs File Delivery
The API returns complete audio files in formats like mp3, wav, opus, flac, or pcm. Partial streaming is not supported, ensuring that each IVR prompt is a discrete, storable asset.
This approach simplifies caching and playback logic for call center operators who manage large libraries of pre-generated messages.
Cost Efficiency Analysis
Pricing is strictly character-based at $10 per million characters. Failed requests incur no charge, protecting your budget during testing phases.
With a minimum top-up of $5 via USDT (TRC-20), teams can test high-volume scenarios without recurring subscription fees or unused seat costs.
Support and Reliability
Support is provided exclusively via email; there is no phone number, live chat, or SLA guarantee. Usage data, including request timestamps and status, is retained for billing purposes only, ensuring transparent cost tracking without storing the generated audio files.
Balance management is straightforward: accounts start with no minimum spend, and the balance never expires. Top-ups via USDT on TRON are credited automatically within minutes, allowing you to monitor consumption directly in your dashboard without waiting for monthly invoices.
Email Support Channel
We do not offer phone support, live chat, or on-site enterprise contracts. All inquiries are handled via email, which is the sole communication channel. This approach keeps overhead low and aligns with our character-based pricing model for indie developers and call center operators.
Usage Monitoring
Request metadata, including time, character count, and status, is kept for billing accuracy. Failed requests are free, and the charge is returned in full if synthesis fails. You can view these metrics directly in the dashboard to predict costs before scaling volume.
- Real-time balance tracking
- Character count per request
- Status logging for failed attempts
Balance Management
Accounts start without trial credit if created via email and password, or with $0.50 if signed in via Google. There are no monthly plans, seats, or minimum spend requirements. Your balance does not expire, so you can top up only when needed.
- No expiration on credit
- Minimum top-up: $5
- Auto-credit via TRON (TRC-20)
Technical Specifications
The API enforces a strict 4,096-character limit per request, counting spaces and punctuation. Longer texts must be split into chunks and joined programmatically.
Speed ranges from 0.25 to 4.0, allowing precise control over generation duration while maintaining audio quality for IVR applications.
Character Limits per Request
Each POST request accepts a maximum of 4,096 characters for the input field. Spaces, punctuation, and numbers count toward this total. For texts exceeding this limit, split the input into multiple requests and concatenate the resulting audio files.
- Maximum input length: 4,096 characters
Speed Adjustment Range
The speed parameter accepts values between 0.25 and 4.0. Adjusting speed affects both playback duration and processing time. Faster speeds may reduce generation latency but alter the voice's natural cadence.
- Minimum speed: 0.25x
- Maximum speed: 4.0x
Response Format Options
Select the output format using the response_format field. Supported formats include mp3, wav, opus, flac, and pcm. Choose formats based on your IVR system's compatibility and storage requirements.
- mp3
- wav
- opus
- flac
- pcm
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. This voice is optimal for general call center announcements where clarity is paramount.
23 languages
Any voice speaks any of these languages.
