Frequently asked questions
What is NewsAudio API?
NewsAudio API is a cloud text-to-speech service designed for editorial teams. You send text via a simple POST request and receive an audio file. It uses standard OpenAI-compatible endpoints, making integration straightforward for newsrooms needing consistent voice output at scale without complex enterprise contracts.
Which voices are best for news delivery?
For news delivery, atlas offers a neutral, news-style voice, while milo provides a friendly, conversational tone. Felix works well for energetic promos, and sage delivers calm, instructional clarity. All eight voices are based on licensed recordings from LibriTTS-R and LibriVox.
How much does it cost per article?
Pricing is $0.01 per 1,000 characters, including spaces and punctuation. Costs depend on your script length. Failed requests are free. There are no monthly plans or minimum spends, so you only pay for the characters you actually synthesize into audio files.
Can I use the API for large teams?
Yes, the API supports large teams efficiently. Each account holds one active API key, which can be issued anytime. With a rate limit of 60 requests per minute, teams can process scripts efficiently. The system handles long texts by splitting them into manageable parts.
Which languages are supported?
We support 23 languages, including English, Spanish, French, German, Japanese, and Arabic. Any voice can speak any supported language. If you do not set the language parameter, the system guesses it from the script, defaulting to English for Latin text. Vietnamese, Thai, Indonesian, Ukrainian, Romanian, Czech, Bengali and others are not supported.
Do you support SSML for emphasis?
No, we do not support SSML. The API accepts plain text input up to 4096 characters. You cannot use phoneme dictionaries or custom pronunciation lexicons. Emphasis is determined by the voice preset and the natural cadence of the generated synthetic speech audio.
Is the audio copyright-free?
The voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain audiobook readings. Each generated file carries an inaudible watermark marking it as synthetic speech. Input text is not used for training, and generated audio is returned directly in the response.
How fast is the generation speed?
Generation is roughly real-time, processing about 25 characters of text per second. Short phrases may take a few seconds. Requests are processed sequentially, so under heavy load, a request might wait in a queue. Long texts are split and joined by the client.
Can I stream the audio in real-time?
No, we do not support streaming or WebSocket connections. The API returns the complete audio file in the response after synthesis is finished. You receive the full MP3, WAV, OPUS, FLAC, or PCM file once generation completes, rather than partial chunks in real-time.
Do you offer a monthly subscription?
No, there are no monthly plans or subscriptions. You pay only for what you use. Top-ups are made via USDT on the TRON network, with a minimum of $5. Your balance does not expire, and there are no seats or minimum spend requirements.
How do I top up my balance?
Top up your balance using USDT on the TRON network (TRC-20). The minimum top-up amount is $5. Credits are added automatically within minutes after your transfer is confirmed. We do not accept credit cards, PayPal, or issue invoices with VAT for these transactions.
What happens if a request fails?
If synthesis fails, the charge is returned in full. You are not billed for failed requests. You can retry the request or split your text into smaller parts if you encounter issues. The system ensures you only pay for successfully generated audio files.
Is my input text used for training?
No, your input text is used only to produce the audio and is not used for training models. Request metadata like time and status is kept for billing purposes. Generated audio is returned in the response and is not stored on our servers.
Do you have a phone number for support?
No, we do not have a phone number or live chat. Support is provided via email only. For technical issues or billing questions, please contact us through the email support channel listed on our contact page.
First request
curl
curl https://api.win9online.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.mp3How NewsAudio API Works
NewsAudio API converts written scripts into audio files through a simple HTTP request. Teams send text to the API endpoint, which processes the input and returns synthesized speech without complex setup.
The system handles character limits by splitting longer articles into parts, ensuring consistent audio quality across entire broadcasts. Generated files are delivered immediately, ready for integration into existing editorial workflows.
Send a script via POST request
Use a standard POST request to send your script. The JSON body accepts text up to 4096 characters, along with parameters for voice, format, and speed. Authentication is handled via a Bearer token in the header.
The API accepts requests in the same shape as the OpenAI speech endpoint, allowing you to use existing OpenAI SDKs by simply changing the base URL. This compatibility reduces integration time for development teams.
- Endpoint: POST https://api.win9online.com/v1/audio/speech
- Body: JSON with 'model', 'input', 'voice', and 'response_format'
- Auth: Bearer token in Authorization header
Receive MP3 or WAV audio files
Upon successful synthesis, the API returns the audio file directly in the response. You can choose between MP3 for web delivery and WAV for high-quality archival needs. Other formats like Opus, FLAC, and PCM are also available.
Each request is processed individually, with typical generation speeds approximating real-time. Short phrases may take a few seconds, while longer texts generate at a consistent pace based on character count.
- Supported formats: mp3, wav, opus, flac, pcm
- Response includes the audio file directly
- Failed requests incur no charge
Handle long articles with splitting
For scripts exceeding 4096 characters, split the text into smaller parts and send multiple requests. The API processes each part sequentially, allowing you to join the resulting audio files into a single continuous broadcast file.
This approach ensures reliable synthesis even for lengthy news articles. Under load, requests may wait in a queue, but processing remains consistent and deterministic for editorial planning.
- Max 4096 characters per request
- Split and join audio files for longer content
- Requests processed sequentially under load
Voices for News and Commentary
NewsAudio API provides eight distinct voice presets designed for editorial clarity and broadcast readiness. Each voice is derived from licensed recordings, ensuring consistent quality across your audio output. The selection allows editorial teams to match the tone of their content precisely, whether delivering hard news or engaging commentary.
Atlas for neutral news delivery
Atlas offers a neutral, professional tone ideal for standard news reading. Its clear enunciation ensures that complex data and breaking updates are communicated without emotional bias,
- Neutral tone suitable for factual reporting
- Clear enunciation for complex data
- Ideal for daily news briefings
Felix for energetic promos
Felix brings an energetic, promo-ready quality to your audio files. Use this voice for segment introductions, sponsor reads, or dynamic updates where a higher energy level helps retain listener attention during short-form content.
- Energetic tone for promos
- Engaging delivery for sponsor reads
Sage for instructional content
Sage provides a calm, instructional voice perfect for educational segments or detailed explainers. Its steady pace aids comprehension, making it suitable for long-form articles, tutorial audio, or contextual background pieces that require listener focus.
- Calm, instructional tone
- Steady pace for comprehension
- Suitable for educational segments

Languages Supported for Global News
NewsAudio API supports 23 languages, ensuring your content reaches global audiences without needing separate voice talent. The service handles script translation needs by allowing direct input in any supported script, with automatic language detection for Latin-based texts to simplify integration.
For non-English Latin text, explicitly setting the language parameter ensures accurate pronunciation. This flexibility allows editorial teams to maintain consistent audio quality across diverse regional broadcasts without managing multiple voice providers.
English, Spanish, French, German
These core languages are fully supported with all eight voice presets, allowing you to choose the perfect tone for your specific segment. Whether delivering straight news or commentary, the voices adapt to your editorial style.
You can generate audio in these languages using any of the available voices, ensuring brand consistency across different language editions of your publication.
- Full voice support for all 8 presets
- High accuracy for major European and American markets
- Consistent tone across news and commentary
Arabic, Chinese, Japanese, Korean
The API handles complex scripts and tonal languages effectively, supporting Arabic, Chinese, Japanese, and Korean. This enables newsrooms to expand their reach into key international markets without hiring specialized voice actors.
Each language leverages the same engine, ensuring uniform audio quality and processing speed regardless of the script complexity.
- Support for complex scripts and tones
- Ideal for expanding into Asian and Middle Eastern markets
- Uniform quality across all language types
23 languages total with auto-detection
The full list includes 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, giving you maximum flexibility.
If the language field is omitted, the system guesses the language based on the script. For non-English Latin text, it defaults to English, so you should explicitly set the language parameter for accurate results.
- 23 supported languages
- Any voice can speak any language
- Auto-detection for Latin scripts (defaults to English)
Formats for Broadcast Integration
NewsAudio API supports multiple audio formats to fit your specific delivery pipeline. You can choose the standard for web playback or the uncompressed standard for professional archiving.
MP3 for web and podcasts
MP3 is the default choice for web embedding and podcast distribution. It offers a balance of small file size and high quality, ensuring fast loading times for listeners while maintaining clear speech intelligibility.
- Default output format
- Ideal for web players and podcasts
- Balanced quality and file size
WAV for high-quality archival
WAV provides uncompressed audio, preserving the full fidelity of the generated voice. This format is ideal for internal editorial archives, broadcast masters, or when downstream processing requires lossless data.
- Uncompressed, lossless quality
- Best for archival and broadcast masters
- Supports standard PCM encoding
Opus, FLAC, and PCM options
For specialized use cases, the API also supports Opus for efficient streaming, FLAC for lossless compression, and raw PCM for direct audio processing. Select the format that matches your technical infrastructure.
- Opus: efficient compression
- FLAC: lossless compression
- PCM: raw audio data
Pricing for Editorial Volumes
NewsAudio API charges $0.01 per 1,000 characters, with spaces and punctuation counting toward the total. This transparent rate allows editorial teams to predict costs accurately without hidden fees or complex tiered structures.
Accounts start with a $0.50 free trial when signing in with Google, providing about 50,000 characters valid for seven days. Top-ups via USDT (TRC-20) are credited automatically, ensuring continuous operation without manual invoicing.
$0.01 per 1,000 characters
The cost is strictly volume-based. You pay only for the characters processed in your requests. Failed requests do not incur charges, and any deducted amount is returned in full if synthesis fails.
- Spaces and punctuation count as characters
- Failed requests are free
- Balance does not expire
No monthly minimums or seats
There are no monthly plans, seats, or minimum spend requirements. You can scale usage up or down freely.
Free failed requests
If synthesis fails, the charge is returned in full. This ensures you only pay for successful audio generation, reducing risk for large batch processing of news scripts.

Integration with Existing Workflows
NewsAudio API integrates directly into your editorial pipelines using standard HTTP requests. The endpoint mirrors the OpenAI speech API structure, allowing you to reuse existing Python or JavaScript clients by simply pointing the base URL to our service.
OpenAI SDK compatible base_url
Configure your SDK client to use https://api.win9online.com/v1 as the base URL. This allows you to use the official OpenAI SDKs without rewriting your integration logic. Authentication uses a standard Bearer token, ensuring seamless drop-in replacement for existing TTS components.
- Set base_url to https://api.win9online.com/v1
- Use standard Bearer token authentication
- Reuse existing OpenAI SDK client code
Simple JSON body structure
Send a JSON payload containing the text, voice selection, and output format. The request structure matches the OpenAI speech endpoint, making it easy for engineers to adapt current scripts. You specify the model, input text, and desired audio format in a single call.
- Standard JSON payload format
- Supports model, input, voice, and format fields
- Easy to parse and generate in any language
No custom SDK required
Since the API follows a widely adopted standard, you do not need to install or maintain a proprietary SDK. Use any HTTP client library to send requests and receive audio files. This reduces dependency overhead and simplifies long-term maintenance for your engineering team.
- No proprietary SDK installation needed
- Works with any standard HTTP client
- Reduces dependency overhead
Speed and Reliability
NewsAudio API generates audio at roughly real-time speed, processing approximately 25 characters of text per second. Long articles are split into parts of up to 4096 characters, allowing editorial teams to receive complete audio files without manual intervention.
Roughly real-time generation
Short phrases take a few seconds to synthesize, while longer scripts generate at a pace that matches spoken delivery. This efficiency ensures that bulk conversions of daily news scripts do not bottleneck your production pipeline.
Queue handling under load
Requests are processed sequentially. Under heavy load, a request may wait in a queue before synthesis begins. This design ensures consistent processing order without requiring complex concurrency management from your side.
Consistent output quality
Each generated file maintains uniform audio characteristics, regardless of length or complexity.
Failed requests are free, so you only pay for successfully delivered audio. This reliability protects your editorial budget during high-volume publishing periods.
Data Privacy for Newsrooms
NewsAudio API is engineered for editorial environments where content integrity matters. When you send a script, the text is processed solely to generate the audio file. It is not ingested into training datasets, ensuring your breaking news stories or exclusive interviews remain proprietary to your organization.
Input text not used for training
Your source text is used exclusively for synthesis. Unlike some consumer models, we do not harvest your scripts to improve our underlying voice engine. This means your unique editorial voice and unpublished stories do not contribute to our training data, keeping your intellectual property secure.
Audio not stored after delivery
Once the audio file is returned in the response, it is not stored on our servers. Each generation is a transient transaction. This minimizes the risk of audio leakage and ensures that your published clips exist only where you choose to host them.
Metadata kept for billing only
We retain minimal request metadata—such as timestamp, character count, and status—strictly for accurate billing and troubleshooting. This data is used to verify your usage against your balance and does not include the content of your scripts or the resulting audio files.

Get Started with a Trial
Create your account at /get-api-key/ using Google for an immediate $0.50 trial credit, valid for seven days. This provides roughly 50,000 characters to test the API without entering payment details. Email-based signups start with zero balance.
Once registered, you receive a unique API key to authenticate requests. The dashboard displays your balance, usage metrics, and a ready-to-use cURL example for immediate testing.
Sign up with Google for trial credit
Select Google Sign-In on the /get-api-key/ page to receive $0.50 in trial credit. This credit expires after seven days. Accounts created with email and password do not receive this initial bonus.
- Instant credit activation
- No credit card required
- 7-day validity period
Generate your API key instantly
Your dashboard provides a single active Bearer key starting with tts-. This key authenticates all API requests. You can issue a new key at any time, which immediately invalidates the previous one.
Test with a curl example
Use the provided cURL command to send your first script. Replace YOUR_API_KEY with your generated key. The endpoint accepts JSON bodies with model, input, and voice parameters.
- Endpoint: POST https://api.win9online.com/v1/audio/speech
- Format: JSON
- Auth: Bearer token
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. The news publication team collaborates across large divisions to ensure accurate and timely reporting.
23 languages
Any voice speaks any of these languages.
Sources and further reading
- Wikipedia
- LibriTTS-R corpus (OpenSLR 141)
- MDN: Web Speech API
- Tether (USDT)
- openai-python on GitHub
- LibriTTS-R paper (arXiv 2305.18802)
- TRON developer hub: TRC-20
- IETF RFC 7845: Ogg Opus
- LibriVox public-domain audiobooks
- Python documentation
- OpenAI API reference: create speech
- W3C WAI: making audio and video accessible
