Frequently asked questions
Is the API suitable for broadcasting news?
Yes. The Atlas voice is a neutral preset, while Orion provides a deep, documentary tone. Both presets are optimized for clear, professional delivery suitable for broadcast workflows. You can generate audio files in formats like MP3 or WAV, which integrate easily into standard CMS or broadcast pipelines without complex setup.
How do I handle articles longer than 4096 characters?
Split your input text into parts of up to 4096 characters each. Send each part as a separate POST request to the /v1/audio/speech endpoint. Then, join the resulting audio files together in your script or workflow. This approach ensures every segment is processed correctly within the character limit per request.
Can I use Python to automate my daily audio newsletter?
Absolutely. Use the official OpenAI Python SDK by changing the base_url to https://api.innwa.info/v1. Your existing Python scripts can send text and receive audio files automatically. This straightforward integration allows you to schedule daily newsletters without building custom infrastructure, leveraging standard HTTP requests and JSON bodies.
What happens if a request fails during generation?
If synthesis fails, the charge is returned in full. You are not billed for failed requests, ensuring you only pay for successful audio generation. This applies to all requests, so you can retry failed calls without worrying about double-charging your account balance or wasting credits on unsuccessful attempts.
Do I need to set the language parameter for English text?
No, it is optional. If you do not set the language parameter, the system guesses it from the script. Latin text defaults to English, so standard English articles will generate correctly without explicit language codes. However, for non-English Latin-script text, you should set the language parameter to ensure accuracy.
Are the voices based on real human recordings?
Yes. The eight voice presets are based on licensed recordings from LibriTTS-R (CC BY 4.0) and public-domain audiobook readings from LibriVox. Each generated file also carries an inaudible watermark marking it as synthetic speech. This ensures synthetic speech derived from real human performances without requiring custom voice cloning.
How long does it take to generate 1,000 characters?
Generation takes roughly real time, processing about 25 characters per second. For 1,000 characters, expect approximately 40 seconds of generation time. Short phrases may take a few seconds longer due to processing overhead. Requests are processed sequentially, so under load, your request may wait in a queue before starting.
Can I switch between different voices in a single script?
Yes. You can specify any of the eight voice presets (Aria, Nova, Luna, Sage, Felix, Milo, Atlas, Orion) in each request. Simply change the 'voice' field in your JSON body for each POST request. This allows you to create dynamic audio content with varied tones within a single automation script.
Is there a limit to how many requests I can make per minute?
Yes, the limit is 60 requests per minute per key. Before your first top-up, the limit is 10 requests per minute. If you exceed this rate, requests may be delayed or rejected. Ensure your automation scripts respect these limits to maintain smooth generation, especially when processing large volumes of editorial text.
Do I need a credit card to sign up?
No. You can sign up using Google or an email and password. Payments are made via USDT on the TRON network (TRC-20). There are no credit cards, PayPal, or subscriptions required. Accounts start with a $0.50 free trial if signed up via Google, or zero balance if using email/password.
What formats are available for the generated audio?
You can choose from MP3, WAV, OPUS, FLAC, and PCM. Specify the desired format in the 'response_format' field of your request. These formats cover most broadcast and web use cases, allowing you to select the best balance of quality and file size for your specific news product workflow.
How do I check my remaining balance in Python?
The dashboard at /get-api-key/ displays your balance, characters left, and usage. While there is no dedicated Python SDK endpoint for balance, you can view this information directly in your account settings. Use the API key credentials shown there to authenticate your requests, and monitor usage via the provided dashboard interface.
Can I use the generated audio commercially?
Yes. The underlying voice recordings are based on LibriTTS-R (CC BY 4.0) and LibriVox public-domain readings. You can use the generated audio in commercial news products, newsletters, or broadcasts. The inaudible watermark identifies the speech as synthetic.
Do you offer refunds if I pay by USDT?
Refunds are automatic only for failed requests, where the charge is returned in full. For successful requests, there is no explicit refund policy mentioned for USDT payments. Credits do not expire, and there are no monthly plans or invoices. Contact support by email for specific billing inquiries regarding USDT transactions.
First request
curl
curl https://api.innwa.info/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 News Teams Choose Python TTS
NewsBrio integrates directly into your CMS via a standard Python API. Send editorial text and receive audio files without complex middleware. This straightforward approach lets product teams focus on content rather than infrastructure setup.
Simple integration with existing CMS workflows
Use the official OpenAI SDK or plain HTTP requests. Change the base URL to our endpoint and you are ready to go. No custom libraries or proprietary wrappers are required.
Reliable batch processing for daily briefings
Split long articles into 4096-character parts and join the resulting audio files. Requests are processed sequentially, ensuring consistent output for your daily news briefings.
Transparent billing without hidden fees
Pay only for what you use. Failed requests are free. Balance never expires, and there are no monthly subscriptions or seat limits to manage.
How It Works
We provide a straightforward Python API that integrates with your CMS or broadcast workflow without complex setup. You send text via standard HTTP requests or the OpenAI SDK by adjusting the base URL to https://api.innwa.info/v1. This approach lets you generate audio files directly from your editorial content using familiar tools.
Send text via Python requests or SDK
Use the standard OpenAI Python SDK or any HTTP client. Change the base_url to our API endpoint and pass your text in the input field. This method works seamlessly with existing codebases, requiring minimal changes to your integration stack.
Receive high-quality MP3, WAV, or OPUS audio
The API returns audio files in your chosen format. You can select MP3, WAV, OPUS, FLAC, or PCM via the response_format parameter. This flexibility ensures compatibility with various broadcasting platforms and content management systems without additional conversion steps.
Handle long articles with automatic splitting
Send texts up to 4,096 characters per request. For longer articles, split the text into parts and join the resulting audio files. This simple workflow ensures consistent quality and manageable processing times for daily briefings or feature stories.
Voice Options for Journalism
Selecting the right voice is critical for maintaining listener trust and engagement in news audio. Our eight presets are designed to match specific editorial tones, from straight news to in-depth features.
Each voice is derived from licensed recordings or public-domain audiobook readings. You can mix and match voices across articles to create a dynamic listening experience.
Neutral voices like Atlas for news delivery
Atlas provides a clear, neutral tone ideal for breaking news and daily briefings. It delivers information without imposing strong emotional bias, making it suitable for factual reporting where clarity is important.
- Neutral, news-ready tone
- Ideal for straight reporting
Friendly tones like Milo for conversational segments
Milo offers a warm, conversational style suitable for interviews, podcasts, or lighter editorial content. His approachable delivery helps humanize the news, making complex stories feel more accessible to listeners.
- Conversational and approachable
- Best for interviews
Deep documentary voices like Orion for features
Orion’s deep, resonant quality suits long-form features and documentary-style audio. His steady pace and rich timbre hold attention during extended narratives, providing a distinct tone to your editorial content.
- Deep, resonant quality
- Great for long-form features
Languages Supported
NewsBrio supports 23 languages, including English, Spanish, French, and German, allowing news teams to produce audio for diverse audiences without managing multiple voice engines. The API accepts text in any of these supported languages, ensuring broad reach for international editorial content.
English, Spanish, French, German, and 19 others
The service covers Arabic, Chinese, Danish, Dutch, Finnish, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Swahili, Swedish, and Turkish. If your content targets languages outside this list, such as Vietnamese or Thai, the API will not generate audio.
- Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish
Automatic language detection or explicit setting
The API guesses the language from the script if you do not specify one, defaulting to English for Latin script. For non-English Latin text, explicitly setting the language field ensures correct pronunciation and avoids misinterpretation.
Consistent quality across all supported languages
All eight voice presets can speak any of the 23 supported languages with consistent quality. This uniformity simplifies your workflow, as you do not need to switch between different voice models or engines when expanding to new markets.
Pricing Transparency
Our pricing is straightforward: $0.01 per 1,000 characters, including spaces and punctuation. Failed requests are free, so you only pay for successful synthesis. This predictable model scales with your editorial volume without hidden fees or per-seat charges.
Start with a $0.50 free trial when you sign in with Google, valid for seven days. No credit card is required to begin testing the API in your newsroom workflow.
Pay per 1,000 characters at $0.01
Each character counts, so long articles are billed proportionally. You can split texts into 4,096-character chunks for reliable processing.
- $0.01 per 1,000 characters
- Failed requests are free
No monthly subscriptions or seat limits
There are no recurring monthly plans or minimum spend requirements. You pay only for what you generate, making it easy to budget for daily briefings or special features.
Top up instantly with USDT on TRON
Add funds via USDT on the TRON network (TRC-20). Minimum top-up is $5, and credits appear automatically within minutes. No invoices, no VAT, and no PayPal.

Technical Reliability
NewsBrio provides a straightforward Python API that integrates with your CMS or broadcast workflow without complex setup. Our infrastructure is designed for reliability, ensuring your daily audio briefings are generated consistently.
With transparent rate limits and predictable performance, news product teams can plan their content pipelines with confidence. We focus on delivering stable, dependable audio generation for editorial workflows.
60 requests per minute rate limit
Each API key supports up to 60 requests per minute. This limit ensures fair usage and consistent performance across all accounts. Before your first top-up, the limit is 10 requests per minute.
- Rate limits are applied per key, not per IP.
- Limits reset automatically every minute.
- Exceeding the limit returns a clear error message.
Real-time generation for short phrases
Short phrases typically generate within a few seconds. This speed allows for quick prototyping and iterative testing in your development environment. For longer texts, generation takes roughly real-time, processing about 25 characters of text per second of audio.
- Short inputs return quickly.
- Long texts process at real-time speed.
Predictable queuing under load
Requests are processed sequentially, so under heavy load, a request may wait in a queue. This predictable behavior means you can estimate processing times accurately. There are no unpredictable spikes in latency during peak usage.
- Sequential processing ensures order.
- Queue wait times are predictable.
Data Privacy & Rights
We prioritize your editorial integrity by ensuring input text is used exclusively to produce the audio file. Your articles are not ingested into our training datasets, preserving the exclusivity of your content strategy.
Text is not used for training models
Your input text serves only to generate the immediate audio output. It is not used to train or improve our underlying models, ensuring your editorial content remains distinct from our training data.
Generated audio is not stored on our servers
Once the audio file is returned in the response, we do not retain a copy of the generated file. Only request metadata—such as timestamp, character count, and status—are kept for billing purposes.
Voices based on licensed and public domain recordings
Our eight voice presets are derived from licensed recordings from LibriTTS-R and public-domain audiobook readings from LibriVox. This foundation provides a legally sound basis for synthetic speech generation in your newsroom workflow.
- Voices carry an inaudible watermark marking them as synthetic speech
Getting Started
Sign up on the /get-api-key/ page using your Google account or a standard email and password. Each account is tied to the specific site you are visiting, ensuring your credentials are isolated and secure.
Once registered, you receive an active API key immediately. Use this key in the Authorization header to start sending text to the API. You can issue a new key at any time, which instantly invalidates the previous one.
Create an account with Google or email
Registration takes seconds. Choose Google for quick access or use your email to create a standard account. This establishes your identity for billing and usage tracking without requiring complex enterprise verification.
Get your API key instantly
Your key appears on the dashboard immediately after sign-up. Copy it into your Python scripts or CMS integrations. The key format starts with 'tts-' and authenticates all requests to the API base.
- One active key per account
- Issue new keys anytime
- Old keys stop immediately
Try the free trial with $0.50 credit
Google sign-ups include a $0.50 trial credit, valid for 7 days. This covers approximately 50,000 characters, allowing you to test integration without immediate payment. Email-based accounts start with zero credit.
- $0.50 for Google sign-ups
- 50,000 characters included
- Valid for 7 days
Support & Contact
Our support team handles all technical inquiries via email, ensuring you receive detailed, thoughtful responses for integration issues or billing questions. We prioritize clear communication to help your newsroom resolve problems efficiently without the noise of live chat or phone queues.
For developers, our documentation provides concrete examples, including ready-to-use Python snippets and curl commands. This resource covers authentication, request structures, and error handling, allowing your team to integrate the API into your CMS or broadcast workflow with minimal friction.
Email support for technical questions
We respond to technical inquiries via email, providing detailed solutions for integration issues, billing discrepancies, or voice selection questions. This asynchronous approach ensures every question receives a thorough, well-reasoned answer from our team.
No live chat or phone support
We do not offer live chat or phone support, keeping our operational overhead low to maintain transparent pricing. You can reach us at [email protected] for any assistance needed with your API usage or account details.
Clear documentation for developers
Our documentation includes ready-to-use Python snippets and curl examples, covering authentication, request structures, and error handling. This resource allows your team to integrate the API into your CMS or broadcast workflow with minimal friction.
- Ready-to-use Python examples
- Curl command templates
- Error handling guides
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. This variation keeps your audio feed dynamic and helps listeners distinguish between different types of content within your broadcast workflow.
23 languages
Any voice speaks any of these languages.
Sources and further reading
- Wikipedia
- IETF RFC 7845: Ogg Opus
- Tether (USDT)
- OpenAI API reference: create speech
- Python documentation
- LibriTTS-R paper (arXiv 2305.18802)
- LibriTTS-R corpus (OpenSLR 141)
- openai-node on GitHub
- LibriVox public-domain audiobooks
- Creative Commons Attribution 4.0
- openai-python on GitHub
- TRON developer hub: TRC-20
