Frequently asked questions
Is the AI voice reader free for commercial IoT products?
No. The service costs $10 per 1 million characters. A free trial provides $0.50 credit for 7 days. Top-up is via USDT on the TRON network (TRC-20) only. There are no monthly plans, seats, or minimum spend requirements. Failed requests are free, but standard usage is billed based on character count.
How many languages does the API support for smart devices?
The API supports 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. Vietnamese, Indonesian, Thai, and others are not supported.
Can I use the free trial credit for production testing?
Yes. The $0.50 trial credit (approx. 50,000 characters) is valid for 7 days after signing in with Google. It can be used for production testing. Accounts created with email and password start without trial credit. Top-up via USDT is required for extended usage beyond the trial period.
What is the maximum text length per API request?
The maximum input text length is 4096 characters per request. For longer texts, you must split the text into parts of up to 4096 characters and join the resulting audio files. Spaces and punctuation count toward the character limit and pricing.
Does the API support streaming audio for real-time interaction?
No. The API returns the complete audio file in the response. There is no streaming, WebSocket, or real-time partial audio support. Requests are processed one after another, so under load a request may wait in a queue.
How are voices selected for different languages?
You select one of eight voice presets (e.g., aria, nova, luna, sage, felix, milo, atlas, orion). Every voice can speak all 23 supported languages. If the language is not set, it is guessed from the script, defaulting to English for Latin text.
Is there a minimum spend requirement for IoT projects?
No. There are no monthly plans, seats, or minimum spend requirements. The minimum top-up amount is $5 via USDT on the TRON network. You pay only for the characters generated, at a rate of $10 per 1 million characters.
What audio formats are available for embedded systems?
The API supports mp3, wav, opus, flac, and pcm response formats. You can specify the desired format in the request body. These formats allow flexibility for various embedded systems and hardware constraints in IoT applications.
Do I need to pay for failed speech synthesis requests?
No. Failed requests are free. If synthesis fails, the charge is returned in full. You are only charged for successful synthesis requests. This ensures you do not pay for errors in your IoT integration workflow.
How long does the free trial credit last?
The free trial credit is valid for 7 days. It is provided when you sign in with Google. The credit amount is $0.50, which allows for approximately 50,000 characters of speech synthesis. Accounts created with email and password do not receive this trial credit.
Can I use the same API key for multiple IoT devices?
Yes. Each account has one active API key that can be used across multiple devices or applications. You can issue a new key at any time, which immediately invalidates the old one. Rate limits are 60 requests per minute per key (10 per minute before the first top-up).
Is the audio output watermarked for synthetic detection?
Yes. Every generated audio file carries an inaudible watermark that marks it as synthetic speech. This helps identify the content as AI-generated voice synthesis without affecting the audible quality or usability in your smart home or IoT products.
What happens if the API rate limit is exceeded?
The rate limit is 60 requests per minute per key (10 per minute before the first top-up). Exceeding this limit results in errors. Requests are processed sequentially, so high load can also cause queuing delays even within limits.
How is the text input processed for different scripts?
Input text is used only to produce audio and is not used for training. If the language is not set, it is guessed from the script. Latin text defaults to English. For non-English Latin-script text, you should explicitly set the language parameter.
First request
curl
curl https://api.play29p.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 NodeVoice for Connected Devices
NodeVoice delivers deterministic pricing for IoT deployments, charging $10 per million characters with no hidden fees or subscription tiers. This predictable cost structure simplifies budgeting for large-scale device networks where voice output volume fluctuates based on user interaction patterns.
Integration relies on standard HTTP endpoints compatible with the OpenAI SDK, allowing seamless integration into existing embedded systems. Developers can switch base URLs to utilize familiar authentication methods without learning proprietary protocols, ensuring quick deployment across diverse hardware architectures.
Deterministic Pricing Models
Pricing is strictly usage-based at $0.01 per 1,000 characters, with spaces and punctuation counted. Failed requests incur no charge, ensuring you only pay for successful synthesis. Accounts start with a $0.50 free trial valid for seven days, eliminating upfront risk for pilot projects.
- Pay only for successful audio generation
- No monthly minimums or seat limits
- Transparent character-based billing
Standard HTTP Integration
The API uses standard RESTful endpoints, accepting JSON bodies for text input and voice selection. Developers can use the official OpenAI SDK by simply changing the base_url to https://api.play29p.com/v1. Authentication is handled via a single Bearer token, simplifying header management in constrained environments.
- RESTful JSON interface
- Compatible with OpenAI SDKs
- Single Bearer token auth
Low Latency for Voice Assistants
Generation speed averages roughly real-time, producing approximately 25 characters of text per second of audio. Short phrases require only a few seconds to synthesize, while longer texts are split into 4,096-character chunks and joined seamlessly. Requests are processed one after another, so under load a request may wait in a queue.
- Avg 25 chars/second generation speed
- Sequential request processing
- Automatic chunking for long texts
Supported Languages and Voices
NodeVoice supports 23 languages, enabling global deployment for smart home and IoT devices. Voices are based on licensed recordings: LibriTTS-R (CC BY 4.0), from LibriVox public-domain audiobook readings.
23 Languages for Global Devices
The API handles Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish. If the language parameter is omitted, the system guesses from the script, defaulting to English for Latin text.
- Supports 23 specific languages for worldwide IoT deployment.
- Automatic language detection for Latin-script inputs.
- No support for Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, or Bengali.
8 Distinct Voice Profiles
Eight presets cover varied use cases, from warm narrators to energetic promo voices. All voices are derived from licensed LibriTTS-R and public-domain LibriVox recordings, carrying an inaudible watermark to identify synthetic speech.
Female options include aria, nova, luna, and sage. Male options include felix, milo, atlas, and orion. Each voice can speak any of the 23 supported languages, allowing flexible configuration for different device types.
- Female: aria, nova, luna, sage.
- Male: felix, milo, atlas, orion.
- All voices support all 23 languages.
Consistent Audio Quality
Audio generation uses deterministic pricing and standard endpoints, ensuring predictable performance for embedded systems. Input text is processed up to 4096 characters per request, with longer texts split and joined seamlessly.
Generated audio is returned in the response and not stored. Request metadata (time, characters, status) is kept for billing.
Technical Specifications
The API follows the OpenAI speech endpoint structure, allowing standard SDKs to function by adjusting the base URL to https://api.play29p.com/v1. This ensures straightforward integration for existing codebases without requiring our own SDK.
OpenAI SDK Compatibility
Requests use the same JSON shape as the official OpenAI speech endpoint. Developers can utilize standard OpenAI SDKs by simply changing the base URL, reducing implementation overhead for IoT firmware and backend services.
Multiple Audio Formats
Output supports mp3, wav, opus, flac, and pcm formats. This flexibility allows edge devices to select the optimal codec for bandwidth constraints or hardware decoding capabilities, ensuring efficient audio delivery.
Character Limit Handling
Each request accepts up to 4096 characters. For longer texts, the client splits the input, synthesizes parts, and joins the resulting audio files. This approach maintains deterministic processing without server-side state management.
Use Cases in IoT
NodeVoice’s deterministic pricing and standard-compliant endpoints enable seamless voice integration into diverse IoT ecosystems. Developers can deploy natural voice responses for hardware without managing complex infrastructure, ensuring consistent audio quality across global devices.
Smart Speaker Responses
Smart speakers benefit from NodeVoice’s 8 voice presets and 23 supported languages. Developers can generate localized responses for user queries, ensuring clear narration for time, weather, or device status updates.
- Support for 23 languages including English, Spanish, and German.
- 8 distinct voice profiles for varied user interfaces.
- Standard HTTP integration via OpenAI SDK compatibility.
Vehicle Infotainment Systems
Vehicle systems require reliable, low-latency audio generation. NodeVoice handles texts up to 4096 characters per request. Longer texts are split into parts and joined.
- Supports MP3, WAV, OPUS, FLAC, and PCM formats.
- Speed adjustable from 0.25 to 4.0 for varied pacing.
- Automatic language detection or explicit ISO code setting.
Industrial Alarm Announcements
Industrial environments need clear, audible alerts. NodeVoice’s neutral voices like Atlas or calm tones like Sage ensure critical messages are understood. Failed requests are free, reducing operational costs for frequent alerts.
- Clear narration options like Atlas (neutral) or Sage (calm).
- No charge for failed synthesis requests.
- Audio returned directly in the API response.
Data Privacy and Storage
Input text is processed exclusively for audio synthesis and is not used for training our models. Generated audio files are returned in the response and are not stored on our servers, ensuring that your device’s spoken content remains transient.
No Training on Input Data
Your raw text inputs are used solely to produce the specific audio file requested. They are not retained in any dataset used to improve or train the underlying speech engine, preserving the privacy of your device’s unique voice commands and responses.
Ephemeral Audio Generation
The synthesized audio is delivered directly in the API response and is not stored for later retrieval or analysis. This ensures that sensitive voice data from smart home devices does not accumulate in long-term storage.
Metadata Retention Policy
Request metadata, including timestamp, character count, and status, is retained strictly for billing and usage verification purposes. This minimal data set supports accurate charging and troubleshooting without retaining the actual content of your voice interactions.

Developer Experience
NodeVoice simplifies IoT integration by leveraging a standard RESTful architecture compatible with existing OpenAI SDKs. Developers can switch base URLs to https://api.play29p.com/v1 and utilize familiar libraries, reducing boilerplate code for embedded systems. This approach ensures consistent behavior across diverse hardware architectures without requiring our own SDK.
Authentication relies on standard Bearer tokens, allowing secure key management within device firmware or backend services. The API returns clear HTTP status codes and structured JSON errors, enabling robust error handling in automated voice generation pipelines for smart home devices.
Simple REST Endpoint
The primary endpoint, POST /v1/audio/speech, accepts standard JSON bodies. This structure mirrors industry standards, allowing seamless integration with existing HTTP clients used in IoT gateways and edge devices. Developers can send text payloads up to 4096 characters and receive audio files directly in the response stream.
- Standard JSON request body
- Compatible with OpenAI SDKs
- Direct audio file response
Standard Authentication
Security is managed via the Authorization header using a Bearer token. Each account holds one active key, which can be rotated instantly via the dashboard. This simplicity reduces key management overhead for devices that need to authenticate with the TTS service without complex OAuth flows.
Clear Error Codes
The API returns standard HTTP status codes to indicate success or failure. Common issues like rate limits (60 requests/minute) or invalid inputs trigger specific error messages. This predictability allows developers to implement reliable retry logic in their smart home applications without guessing error states.
Pricing Transparency
NodeVoice employs a deterministic pricing model tailored for IoT scale. Costs are calculated strictly at $0.01 per 1,000 characters, including spaces and punctuation. This predictable rate simplifies budgeting for high-volume device deployments.
Pay Per Character
Usage is charged per character processed, with failed requests refunded in full. This ensures you only pay for successful audio synthesis, aligning costs directly with usable output.
- $0.01 per 1,000 characters
- Failed requests are free
- Balances do not expire
No Hidden Fees
There are no monthly plans, seat fees, or minimum spend requirements. Pricing remains consistent regardless of volume, eliminating surprise charges for edge devices or burst traffic.
Free Trial Credit
New accounts signed in with Google receive $0.50 in trial credit (approx. 50,000 characters), valid for 7 days. Email-based accounts start with zero credit.
- $0.50 trial credit for Google sign-ins
- Approx. 50,000 characters
- Valid for 7 days
Integration Workflow
NodeVoice integrates into IoT ecosystems via standard HTTP requests, ensuring compatibility with existing embedded frameworks. Developers generate a unique API key through the /get-api-key/ dashboard, which authorizes all subsequent speech synthesis requests.
API Key Generation
Access the dashboard to create a single active key per account. Revoke or regenerate keys instantly via the interface; the previous key becomes invalid immediately upon issuance.
Request Formatting
Send a POST request to https://api.play29p.com/v1/audio/speech with a JSON body. Include the Authorization: Bearer header, model (tts-1 or tts-1-hd), input text, and voice preset. Ensure text does not exceed 4096 characters per request.
Audio Delivery
The API returns the audio file directly in the response body. Supported formats include mp3, wav, opus, flac, and pcm. For texts exceeding the character limit, split input into chunks and concatenate the returned audio files.
Support and Documentation
NodeVoice provides direct email support for technical inquiries, ensuring developers receive clear, written responses without the noise of live chat or phone queues. This channel handles integration questions, billing discrepancies, and voice-specific queries for smart home deployments.
Email Support Channel
Our support team responds to all submitted tickets via email, maintaining a transparent and asynchronous communication style suitable for distributed development teams. There are no phone numbers or live chat widgets; all assistance is documented and searchable.
Comprehensive Docs
The documentation covers the full REST API specification, including endpoint definitions, parameter descriptions, and error code explanations. Developers can find ready-to-use curl examples and integration guides for the OpenAI-compatible SDK setup directly on the /docs/ page.
- Full API reference for /v1/audio/speech and /v1/voices
Community Resources
We do not maintain a formal forum. Support is by email only. This allows you to leverage existing community knowledge for common integration issues.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. This creates a more consistent user experience across different devices within a smart home or industrial network without manual reconfiguration.
23 languages
Any voice speaks any of these languages.
