Frequently asked questions
Who runs Podcast Voice Lab?
Podcast Voice Lab is an independent storefront for a cloud text-to-speech API. We operate as a standalone service, not a division of a larger tech company. Our team manages the API infrastructure and customer support directly via email. We are not affiliated with OpenAI, ElevenLabs, or any other major brand mentioned in industry comparisons.
Are you affiliated with OpenAI?
No, we are not affiliated with OpenAI. Our API endpoint mimics the OpenAI speech endpoint structure, so you can use official OpenAI SDKs by simply changing the base_url to https://api.playli789.com/v1. We use the same model names (tts-1, tts-1-hd) but run on our own independent infrastructure.
Where do the voices come from?
Our eight voices are based on licensed recordings from LibriTTS-R, which uses data from LibriVox public-domain audiobook readings. Each voice is a distinct preset, such as Aria or Felix. The audio files generated by our API carry an inaudible watermark to mark them as synthetic speech.
Is the CC BY 4.0 license commercial?
Yes, the CC BY 4.0 license allows commercial use. Since our voices originate from LibriVox public-domain readings, you can use the generated audio for commercial podcasts, ads, and shows without additional licensing fees for the voice source itself, provided you comply with our API terms.
Do you sell user data?
No, we do not sell user data. 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) strictly for billing purposes.
How long do you keep request logs?
We keep request metadata, including time, character count, and status, for billing purposes. This ensures accurate charges based on the $0.01 per 1,000 characters rate. We do not retain the actual input text or generated audio files after the request is processed.
Can I request data deletion?
While we do not store input text or audio files, we retain request metadata for billing. You can contact us via email to discuss your specific data retention needs. Since we do not sell data or use text for training, your privacy is maintained by design.
Do you have a physical address?
As an independent API provider, we operate digitally. Support is available exclusively via email. We do not have a physical storefront or office open to the public, keeping our operations lean and focused on API delivery.
Who is the editorial team?
Podcast Voice Lab is run by an independent team managing the API and support. Our focus is on providing reliable TTS infrastructure for podcast publishers, with support handled directly through our email channel.
Are the voices real people?
The voices are based on real human recordings from LibriVox, but they are synthesized presets. They are not live performances by specific individuals. Each generated file includes an inaudible watermark to identify it as synthetic speech, distinguishing it from traditional human-recorded audio.
Do you use AI to generate voices?
We use a cloud text-to-speech engine to convert text into audio. The voices are presets derived from licensed recordings. We do not use generative AI to create new voice personas on the fly; we offer eight distinct, human-sounding presets like Aria and Felix.
Is the API open source?
No, our API is not open source. We provide a standard REST API endpoint compatible with OpenAI SDKs. You interact with our service via HTTP requests and JSON bodies, using our API keys.
Can I use the API for books?
Yes, you can use the API for books. You can send text up to 4096 characters per request. For longer texts, split them into parts and join the audio files. This workflow is suitable for audiobook production, though you must manage the splitting logic.
Do you offer refunds?
Failed requests are free; if synthesis fails, the charge is returned in full. You can contact us via email regarding billing issues. We do not offer automatic refunds for successful generations, only for failed ones.
Who We Are
Podcast Voice Lab is an independent storefront for a text-to-speech API, designed specifically for podcast publishers. We provide a straightforward way to generate professional narration from text, eliminating the need to hire voice actors for every script.
Our editorial team focuses on delivering clear, honest tools for content creators. We believe that high-quality audio should be accessible without the overhead of traditional recording sessions, allowing publishers to scale their production efficiently.
Editorial Team
We are a small, dedicated group of editors and engineers who understand the nuances of podcast production. Our background in publishing ensures that our API reflects the needs of modern audio storytellers.
Independent Storefront
Unlike large tech giants, we operate as an independent storefront. This means we control the entire experience, from the API endpoint to your billing, ensuring a direct relationship with our users.
Mission
Our mission is to empower podcast publishers with reliable, human-sounding voices. We aim to simplify the production workflow, letting you focus on content rather than scheduling studio time or managing complex audio files.
Voice Licensing
We source our eight voice presets from licensed recordings in the LibriTTS-R dataset. This foundation ensures consistent, high-quality narration suitable for podcast intros, ads, and long-form episodes without requiring custom voice actors for every script.
LibriTTS-R Source
Our voices are based on licensed recordings from the LibriTTS-R dataset. This collection provides a robust base for generating natural-sounding speech across multiple languages and accents.
CC BY 4.0 License
The underlying recordings are derived from LibriVox public-domain audiobook readings. This licensing model allows us to offer professional-grade narration while respecting the original creators' rights under Creative Commons Attribution 4.0.
Voice Origins
Each voice preset, from Aria’s warm narration to Orion’s deep documentary tone, is built from these curated recordings. We do not use celebrity imitations or custom cloning; every generated file carries an inaudible watermark marking it as synthetic speech.
All eight voices can speak any of the 23 supported languages, ensuring versatility for global podcast publishers.
- Aria: warm, clear narrator
- Felix: energetic, promo style
- milo: friendly, conversational

Data Policy
We process your input text solely to generate the requested audio file. Your scripts are not used to train our models or improve the voice engine, ensuring your original content remains yours.
Generated audio is returned in the response and not stored on our servers. We retain only request metadata—such as timestamp, character count, and status—for billing purposes, keeping your data usage precise and transparent.
Input Text Usage
Text sent to the API is used exclusively for synthesis. We do not repurpose your input for training or any other purpose, keeping your script data strictly functional.
Audio Storage
We do not store the resulting audio files after delivery. Each request is processed, the file is returned to you, and then it is removed from our active memory.
Metadata Retention
We keep minimal metadata: time, character count, and status. This data is retained solely to support accurate billing and to verify usage limits for your account.
Product Facts
Our API follows the standard OpenAI request shape, allowing you to integrate quickly using existing SDKs by simply changing the base URL. You send text and receive an audio file via a simple POST request.
API Specifications
Send text up to 4,096 characters per request. You can choose from five audio formats (mp3, wav, opus, flac, pcm) and adjust speed between 0.25x and 4.0x. The system guesses the language from your text, though you can specify it explicitly.
Rate Limits
New accounts start with a limit of 10 requests per minute. Once you top up your balance, this increases to 60 requests per minute. Requests are processed sequentially, so complex generations may wait in a queue during high load.
Model Versions
We offer two models: "tts-1" and "tts-1-hd". Both use the same underlying engine and pricing, giving you the flexibility to test quality without changing your integration code.

Support and Contact
Podcast Voice Lab is an independent storefront, not a large corporation. We support podcast publishers directly, focusing on clear communication for API integration and billing questions.
Email Support
All support requests are handled via email. We do not offer phone lines or live chat, keeping our operations lean and focused on the API product.
No Live Chat
We avoid live chat to ensure every query receives a thoughtful, written response that can reference specific API logs or billing details without the pressure of real-time typing.
Response Time
We aim to reply within 24 hours on business days. During high traffic, responses may take slightly longer, but we prioritize clarity over speed.
- Email: [email protected]
- Business days: Monday–Friday
- Average response: <24 hours

Transparency
We believe transparency is essential for publishers who integrate voice technology into their workflows. Our pricing is straightforward, our limits are explicit, and our claims are grounded in measurable facts rather than marketing hype.
No Hidden Fees
You pay exactly $0.01 per 1,000 characters, with spaces and punctuation counted. There are no monthly subscriptions, no per-seat charges, and no VAT invoices. Failed requests are free, and your balance never expires.
Clear Limits
Requests are limited to 60 per minute per key. Each text input is capped at 4,096 characters. We do not guarantee specific uptime or latency, as processing queues can vary under load.
Honest Claims
We do not claim to be the 'best' or 'fastest' TTS provider. Our voices are based on licensed LibriTTS-R recordings, and we clearly distinguish our API from OpenAI’s ecosystem. We avoid superlatives and stick to verifiable facts.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Captivating audio storytelling transforms passive listeners into devoted subscribers, driving sustained growth for forward-thinking publishing houses today.
23 languages
Any voice speaks any of these languages.
