Frequently asked questions
How does the text-to-speech AI work for podcast scripts?
Send your script via POST to /v1/audio/speech. The API returns audio using the tts-1 or tts-1-hd models. Specify voice, language, and speed. Requests are processed sequentially, so longer scripts may queue. Input text is used only to produce audio and is not used for training.
Can I use the API to generate full episode audio?
Yes. Split texts exceeding 4096 characters into parts, generate each separately, and join the audio files. Generation is roughly real-time (25 characters per second). Requests process sequentially, so under load, they may queue. This workflow suits episodic content pipelines where consistency matters more than instant delivery.
Which voices are best for narrative podcasts?
Luna offers a soft, storytelling tone, while Aria provides a warm, clear narrator style. Sage is suitable for instructional segments. These voices are based on licensed recordings and can speak any of our 23 supported languages. Adjust the speed between 0.25 and 4.0 to match your podcast's pacing requirements.
How many languages are supported?
The API supports 23 languages, including English, Spanish, German, French, Japanese, and Arabic. Any voice can speak any supported language. If the language parameter is omitted, the system guesses the language from the script, defaulting to English for Latin text. Notable unsupported languages include Vietnamese, Thai, Ukrainian, Indonesian, and Romanian.
Does the API support different audio formats like MP3 or WAV?
Yes. Specify response_format as mp3, wav, opus, flac, or pcm. These formats are available for all voice models. Every generated file contains an inaudible watermark marking it as synthetic speech. Failed requests are free: if synthesis fails, the charge is returned in full, ensuring you only pay for successful generations.
What happens if my script exceeds 4096 characters?
Each request accepts up to 4096 characters. You must split longer scripts into smaller parts, generate audio for each part individually, then join the audio files to create the final episode. This ensures reliable processing. The API does not automatically split text; your application handles the chunking logic.
Is there a limit to how many requests I can make?
Your API key is limited to 60 requests per minute. New accounts without a top-up are restricted to 10 requests per minute. Exceeding these limits causes requests to queue or fail. Each account holds one active key, which you can rotate at any time to invalidate the old one. Monitor your usage and limits directly via your dashboard.
Are the generated voices copyrighted?
Our voices are synthesized from licensed LibriTTS-R (CC BY 4.0) and public-domain LibriVox readings. Each generated file includes an inaudible watermark marking it as synthetic speech. Your input text is not used for training. We do not store generated audio; it is returned directly in the API response. You receive the file immediately for your use.
How is the price calculated per character?
Pricing is $10 per 1 million characters ($0.01 per 1,000 characters). Spaces and punctuation count as characters. Failed requests are free with full charge returns. There are no monthly plans, seats, or minimum spends. Your balance never expires. Top up via USDT on the TRON network (minimum $5), credited automatically within minutes.
Can I use the audio for commercial podcast releases?
Yes. Our API delivers broadcast-ready audio designed for episodic content pipelines. Voices are based on licensed recordings and public-domain audiobook readings. You pay only for the characters generated; there are no monthly plans, subscriptions, or additional licensing fees. Input text is used solely for synthesis, and generated audio is returned in the response without being stored.
Does the API support streaming audio?
No. The API returns the complete audio file only after synthesis finishes; it does not support streaming, WebSocket, or real-time partial audio. Requests are processed sequentially, so longer texts may take time, while short phrases typically take a few seconds. You receive the full file via the standard HTTP response, ready for immediate download and use in your podcast workflow.
What is the response time for short phrases?
Short phrases typically take a few seconds to generate. Longer texts process at roughly real-time speed, approximately 25 characters per second. Requests are processed sequentially, so under load, a request may wait in a queue. If synthesis fails, the charge is returned in full.
Do I need to install any software to use this API?
No installation is required. Use the OpenAI SDKs with base_url https://api.playmm33.com/v1 or send plain HTTP POST requests to /v1/audio/speech. You only need your API key and a JSON-capable client. We do not provide a mobile SDK or WordPress plugin, so integration relies on standard web protocols and code.
How do I top up my balance?
Top up using USDT (TRC-20) on the TRON network. The minimum deposit is $5, and funds are credited automatically within minutes. We do not accept credit cards, PayPal, or subscriptions. Once you sign up via Google or email, you can manage your API key and balance directly in the dashboard.
First request
curl
curl https://api.playmm33.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.mp3Engineered for Podcast Workflows
EchoCast is a cloud TTS API designed for podcasters who prefer code over drag-and-drop tools. We provide a reliable engine that converts scripts into broadcast-ready audio, ideal for episodic content pipelines.
Our API processes requests sequentially, delivering consistent synthetic voices with an inaudible watermark. This ensures your audio remains safe and your production pipeline stays automated without manual intervention.
Automated Script-to-Audio Pipelines
Send text via POST to our endpoint and receive an audio file. With up to 4096 characters per request, you can automate long-form script generation by splitting text and joining the resulting audio files.
Consistent Voice Across Episodes
Choose from eight presets like Aria or Atlas to maintain a uniform tone across your series. Each generated file carries an inaudible watermark, marking it as synthetic speech while preserving brand identity.
No Subscription Lock-in
Use a pay-per-character model with no monthly plans or minimum spend. Top up via USTRC-20 and keep your balance. One active key per account, issued instantly upon sign-in.
Voice Presets for Every Genre
EchoCast provides eight distinct voice presets tailored for podcast production, ranging from warm narrators to neutral news readers. Each voice delivers consistent tone and pacing, ensuring your podcast sounds uniform across all episodes without the variance of human re-recording. All voices support the same 23 languages and can be adjusted for speed to match your script’s pacing.
Narrative and Storytelling Voices
Aria offers a warm, clear narrator style ideal for audiobooks and episodic storytelling. Luna provides a soft, engaging tone suited for long-form narrative content. Both ensure consistent delivery for serialized shows.
- Aria: warm, clear narrator
- Luna: soft, storytelling
News and Documentary Styles
Atlas delivers a neutral, professional voice perfect for news updates or factual content. Orion offers a deep, documentary-style tone for investigative pieces. These voices maintain authority and clarity across episodes.
- Atlas: neutral, news
- Orion: deep, documentary
Conversational and Promo Tones
Milo provides a friendly, conversational tone for interviews or casual segments. Felix brings energetic delivery suitable for promotions or calls to action. Nova adds a bright, upbeat quality for lighthearted segments. Sage offers a calm, instructional style for educational content.
- Milo: friendly, conversational
- Felix: energetic, promo
- Nova: bright, upbeat
- Sage: calm, instructional

Global Language Support
Our engine supports 23 languages, covering major markets from Arabic to Swahili. Every voice preset can speak any supported language, allowing you to produce consistent episodic content across diverse audiences without managing separate voice models.
The system automatically detects language from the text script. For non-English Latin-script text, you must explicitly set the language code to ensure accuracy. Languages like Vietnamese, Thai, and Ukrainian are not supported.
23 Supported Languages
We support Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish.
Automatic Language Detection
If you do not specify a language, the API guesses it from the script. Latin script defaults to English.
This works well for standard English text but may misidentify other Latin-based languages.
Explicit Language Configuration
Send the language parameter with an ISO code to force the correct voice output.
Use this for any non-English text in Latin script to prevent mispronunciation.
Voices are not restricted by language; any voice preset can deliver any supported language.
Simple, Transparent Pricing
Pricing is straightforward: $0.01 per 1,000 characters, including spaces and punctuation. Failed requests are free, with the charge returned in full. You pay only for what you generate, with no monthly subscription lock-in or hidden fees.
Pay Per Character Used
The rate is $10 per 1 million characters. This flat rate applies to both tts-1 and tts-1-hd models, which use the same engine.
No Minimum Spend
There is no minimum spend, no monthly subscription lock-in, and no hidden fees. Top-ups via USDT (TRC-20) start at a $5 minimum, credited automatically within minutes.
Free Trials for New Accounts
Sign up with Google to receive $0.50 of credit, valid for 7 days. This provides about 50,000 characters to test the API. Email-based accounts start without trial credit.
Technical Specifications
EchoCast provides a straightforward API endpoint that integrates directly into podcast production pipelines. By utilizing the standard POST /v1/audio/speech endpoint, developers can leverage existing OpenAI SDKs with a simple base URL change. This ensures consistent behavior and reduces integration time for engineering teams managing audio workflows.
The API accepts text input up to 4096 characters, supporting multiple audio formats including mp3, wav, opus, flac, and pcm. This flexibility allows users to choose lossless formats like WAV or FLAC for archival quality, or efficient formats like MP3 and Opus for web delivery, depending on their specific storage and processing needs.
OpenAI-Compatible Endpoint
Use the standard POST /v1/audio/speech endpoint with JSON body structure. Authentication is handled via Bearer token in the Authorization header. The endpoint accepts model selection (tts-1 or tts-1-hd), input text, voice preset, response format, and optional speed and language parameters. This compatibility means you can switch from other providers by simply updating your base URL and API key.
- POST /v1/audio/speech endpoint
- JSON body with standard fields
- Bearer token authentication
Multiple Audio Formats
Choose from mp3, wav, opus, flac, or pcm formats. MP3 offers efficient storage and broad web player compatibility. WAV and FLAC provide lossless quality suitable for archival or professional editing. Opus and PCM offer further flexibility for different processing needs, ensuring the audio file meets your specific pipeline requirements.
- mp3: Efficient storage, web-ready
- wav/flac: Lossless archival quality
- opus/pcm: Flexible processing
Real-Time Generation Speed
Generation is roughly real-time, producing about 25 characters of text per second of audio. Short phrases may take a few seconds to process. For longer scripts, split the text into parts of up to 4096 characters and join the resulting audio files. Requests are processed sequentially, so under load, a request may wait in a queue before synthesis begins.
- ~25 characters per second of audio
- 4096 character max per request
- Sequential processing under load

Audio Quality and Rights
Audio quality is derived from licensed recordings of public-domain audiobooks, ensuring a natural, professional tone suitable for long-form podcasting without the uncanny valley effect. Our engine maintains consistent voice characteristics across episodes, which is critical for episodic content pipelines where listeners expect uniformity.
Licensed Voice Recordings
Our eight voice presets are based on licensed recordings from the LibriTTS-R dataset (CC BY 4.0), sourced from LibriVox public-domain audiobook readings. This foundation ensures that the synthetic output maintains a natural, professional tone suitable for long-form podcasting.
- Based on LibriTTS-R (CC BY 4.0)
Inaudible Synthetic Watermark
Every generated audio file carries an inaudible watermark that marks it as synthetic speech. This allows podcasters and platforms to easily identify AI-generated content, ensuring transparency with your audience without affecting the listening experience.
No Training on Your Data
Your input text is used solely to produce the audio response and is not used for model training. Generated audio is returned in the response and not stored, ensuring your scripts and voice outputs remain private and secure.
Integration Overview
The API accepts standard JSON payloads via HTTP POST, making it easy to integrate into existing podcast production pipelines. You can test connectivity immediately using command-line tools or your preferred HTTP client without requiring custom drivers.
For developers, the endpoint is compatible with official OpenAI SDKs by adjusting the base URL. This allows you to use familiar Python or JavaScript libraries to handle authentication, request formatting, and audio retrieval with minimal code changes.
cURL and HTTP Requests
Send a POST request to /v1/audio/speech with a JSON body containing model, input, voice, response_format, and optional speed or language parameters. Authentication is handled via a Bearer token in the Authorization header.
- Endpoint: POST https://api.playmm33.com/v1/audio/speech
- Auth: Authorization: Bearer tts-...
Python and JS SDK Support
The official OpenAI SDKs work seamlessly with EchoCast by simply pointing the base_url to our API domain. This allows you to use familiar Python or JavaScript libraries to handle authentication, request formatting, and audio retrieval with minimal code changes.
Batch Processing for Long Scripts
Since each request supports up to 4096 characters, long podcast scripts should be split into chunks before submission. Your application should join the resulting audio files sequentially to create a continuous broadcast-ready file.
Why Podcasters Choose Us
We avoid proprietary editors and drag-and-drop interfaces. Your script lives in your codebase, not in our platform. This keeps your production pipeline simple and version-controlled.
No Proprietary Ecosystem
Unlike platforms that lock you into their editors, we provide a standard cloud API. You control the workflow, not the tool.
Direct API Access
Access our voices directly via REST endpoints. Use the official OpenAI SDKs or plain HTTP requests. Change the base URL and start generating audio immediately.
- Standard JSON request shape
- Works with existing OpenAI clients
- No custom SDK required
Engineering-First Interface
We prioritize technical clarity and reliability. The interface is designed for developers who prefer code over UI. Get clear responses for every request.

Get Started
Begin with a $0.50 free trial valid for 7 days when signing in with Google. This credit covers approximately 50,000 characters, allowing immediate validation of voice quality and integration speed without financial commitment.
Access your unique API key from the dashboard at /get-api-key/. The interface displays your active key, current balance, and character usage, providing all necessary credentials to begin production.
Sign Up for Free Trial
Create an account using your Google credentials to instantly receive $0.50 in trial credit. This amount allows you to generate roughly 50,000 characters of audio over the 7-day validity period, sufficient for testing multiple voice presets.
Generate Your API Key
Navigate to the dashboard to view your active Bearer token and current balance. One key is issued per account; you can regenerate it at any time, with the old key stopping immediately.
First Audio Request
Send a POST request to https://api.playmm33.com/v1/audio/speech. Include your key in the Authorization header and select a model like tts-1. The API returns the audio file directly, ready for insertion into your podcast workflow.
- Use model: tts-1 or tts-1-hd
- Set voice: aria, nova, luna, sage, felix, milo, atlas, or orion
- Format: mp3, wav, opus, flac, or pcm
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Stop coding and start creating your dream app with just your voice and a few simple taps.
23 languages
Any voice speaks any of these languages.
Sources and further reading
- Wikipedia
- Creative Commons Attribution 4.0
- LibriTTS-R corpus (OpenSLR 141)
- IETF RFC 7845: Ogg Opus
- Tether (USDT)
- curl documentation
- W3C WAI: making audio and video accessible
- OpenAI API reference: create speech
- openai-node on GitHub
- Node.js documentation
- LibriVox public-domain audiobooks
- LibriTTS-R paper (arXiv 2305.18802)
