Frequently asked questions
Which voice is best for news?
Atlas is designed for neutral, news-style delivery. For conversational segments, Milo offers a friendly tone, while Felix suits energetic promos. Any of the eight voices can read any supported language. Choose based on whether you need authoritative reporting or engaging storytelling for your editorial content.
Is NewsAudio better than competitors?
We are not affiliated with OpenAI, ElevenLabs, Google, Amazon, Microsoft or any other brand mentioned in comparisons. Our pricing is $10 per million characters with no monthly plans. We offer USDT (TRC-20) top-ups and a free trial. Unlike many competitors, we do not require enterprise contracts or minimum spends, keeping billing transparent and simple.
Do I need a developer?
A developer helps integrate the API efficiently, but it is not strictly required. The endpoint matches the OpenAI speech format, so existing OpenAI SDKs work by changing the base URL. You can also use the provided curl examples or standard HTTP requests to generate audio files.
Can I use it for podcasts?
Yes. Luna provides a soft, storytelling voice ideal for narrative podcasts. Sage offers a calm, instructional tone suitable for educational content. You can adjust speed between 0.25 and 4.0 to match your pacing needs. The API handles up to 4096 characters per request; longer texts must be split into parts and the audio files joined.
What is the character limit?
Each request accepts up to 4096 characters of input text. Spaces and punctuation count toward this limit. For longer scripts, split the text into parts and join the resulting audio files. This ensures consistent quality and manageable processing times for your broadcast-ready audio production workflow.
Do you support SSML?
No, we do not support SSML. The API uses simple text input with optional language codes. You cannot add phoneme dictionaries or custom pronunciation lexicons. Adjust timing and tone by changing the voice preset or modifying the speed parameter between 0.25 and 4.0.
Is the audio copyright-free?
Voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain readings. Generated files carry an inaudible synthetic watermark. Input text is not used for training. We do not offer voice cloning or custom rights management, so verify specific usage rights for your distribution channel.
How fast is the generation?
Generation is roughly real-time, processing about 25 characters per second. Short phrases take a few seconds; long texts may queue under load. Requests are processed sequentially. Expect faster turnaround for shorter clips. Failed requests are free and do not consume your character balance or time.
Voice Quality and Style
The NewsAudio API provides eight distinct voice presets designed to match the specific tone of your editorial content. Whether you need the neutral, authoritative delivery of Atlas for hard news or the bright, upbeat energy of Nova for lifestyle segments, you can select the voice that best fits the script. Each voice is based on licensed recordings, ensuring professional audio quality from the first generation.
Neutral vs. energetic voices
Choose between calm, instructional tones like Sage or energetic promo styles like Felix. This range allows editorial teams to maintain brand consistency while adapting to different content types, from serious political updates to lighter weekend features.
- Atlas for neutral, news-style delivery
- Nova for bright, upbeat segments
- Sage for calm, instructional content
- Felix for energetic, promo-style reads
Clarity for news delivery
All voices are optimized for clear articulation, ensuring that complex names, dates, and financial figures are pronounced correctly. The engine handles mixed-language scripts by guessing the language from the script, though setting the language explicitly is recommended for non-English Latin text to ensure accuracy.
Consistency across episodes
Use the same voice preset across daily briefs or weekly podcasts to create a recognizable audio identity. The API returns consistent audio files for identical inputs, allowing teams to standardize their output without worrying about variability in voice tone or speed between episodes.
Language Support for Global News
The NewsAudio API supports 23 languages, enabling newsrooms to produce content for diverse audiences without managing multiple voice providers. This range covers major global markets, ensuring your broadcasts reach listeners in their preferred language with consistent quality.
For workflows where the input text uses Latin script, the API automatically detects the language. However, for non-English Latin text, you must explicitly set the language code to ensure accurate pronunciation and avoid defaulting to English.
23 languages supported
Our 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. Any voice preset can speak any of these languages, allowing you to maintain a consistent brand voice across different linguistic regions.
- Arabic, Chinese, Danish, Dutch, English, Finnish, French, German
- Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian
- Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish
Auto-detection for Latin script
If you do not specify a language, the system guesses it from the script. Text in Latin characters defaults to English. This simplifies integration for English-first newsrooms but requires care for multilingual content.
Explicit language setting
For non-English Latin-script text, always set the language field. Without this, the API may misinterpret words or apply English phonetics. Not supported: Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, Bengali and all others not in the list.

Pricing and Cost Efficiency
NewsAudio API charges $0.01 per 1,000 characters, including spaces and punctuation. This flat rate applies to both tts-1 and tts-1-hd models, ensuring predictable costs for daily news briefs or long-form article narrations. With a free trial of $0.50, editorial teams can test volume without upfront commitment.
Billing uses USDT (TRC-20) with no monthly minimums, subscriptions, or VAT invoices. Failed requests are refunded in full, so you only pay for successful synthesis. Balances do not expire, allowing newsrooms to top up only when needed, avoiding idle subscription costs while maintaining consistent audio output.
Character-based pricing
Each request is charged based on the exact character count of the input text. Spaces and punctuation are included in this count. The rate is $0.01 per 1,000 characters for both standard and high-definition models.
Long articles may require splitting into segments under 4,096 characters. Each segment is billed separately. This granular billing ensures you only pay for the audio actually generated and delivered successfully.
- $0.01 per 1,000 characters
- Spaces and punctuation count as characters
- Same price for tts-1 and tts-1-hd
No monthly minimums
There are no monthly plans, seats, or minimum spend requirements. You can scale usage up or down based on editorial needs without penalty.
Accounts start with a $0.50 free trial valid for 7 days. Email-based accounts do not receive this trial credit. Top-ups are made via USDT on the TRON network, credited automatically within minutes.
No invoices with VAT are issued. The system is designed for flexibility, allowing news teams to manage budgets directly without enterprise contracts or fixed recurring fees.
Free failed requests
If synthesis fails due to server issues or invalid parameters, the charge is returned in full. You are not billed for audio that is not successfully delivered.
This ensures cost efficiency during peak load times or when testing new voices. Failed requests do not consume your balance, allowing for accurate budgeting based on actual usage.
Integration Ease
The API is designed for immediate adoption by engineering teams already familiar with modern speech services. By mirroring the request shape of the OpenAI speech endpoint, you can integrate NewsAudio into existing workflows with minimal code changes. Simply update the base_url to https://api.win9online.com/v1 and supply your unique API key in the authorization header.
OpenAI SDK compatibility
Official OpenAI SDKs work seamlessly without modification. This eliminates the need for custom wrappers or learning a new library. You can use the same Python, Node.js, or Go clients your team already trusts, ensuring a smooth transition from testing to production environments.
- Compatible with standard OpenAI client libraries
- Supports Python, JavaScript, Go, and other major languages
- Identical request structure to OpenAI's speech endpoint
Simple JSON structure
Requests are standard JSON payloads sent via POST. You provide the model, input text, voice, and optional parameters like speed or format. The response returns the audio file directly, allowing your application to handle the binary data or save it to storage instantly.
- Standard JSON request body
- Direct audio file return in response
- Easy to parse with any HTTP client
No custom SDK required
You do not need to install a proprietary software development kit. Basic HTTP requests are sufficient for all operations, from generating speech to retrieving available voices. This keeps your dependency tree light and reduces potential version conflicts in your newsroom's tech stack.
Speed and Reliability
Newsrooms need predictable audio production timelines. Our API generates audio at roughly real-time speed, meaning a one-minute news segment takes about one minute to process. This allows editorial teams to plan batch conversions for daily briefs without unexpected delays.
For shorter phrases, generation may take a few seconds. Under heavy load, requests are processed sequentially, so your job may wait in a queue. This ensures consistent engine performance without compromising the integrity of the broadcast-ready audio files.
Real-time generation speed
The engine processes approximately 25 characters of text per second of generated audio. For a standard 60-second news clip, you should expect a processing time of roughly 60 seconds. This linear relationship makes it easy to estimate turnaround times for bulk article conversions.
- Roughly 25 characters per second of audio
- Short phrases may take a few seconds
- Predictable timing for batch scheduling
Queue handling under load
Requests are processed one after another. If multiple team members trigger synthesis simultaneously, additional requests will wait in a queue until the previous task completes. This sequential processing prevents engine overload and maintains stable performance during peak news hours.
- Sequential processing model
- Requests queue automatically under load
- No priority tiers or paid acceleration
Consistent output quality
Each audio file is generated with the same voice preset parameters. Whether you are producing a morning briefing or an evening update, the tone remains uniform. This consistency is critical for editorial teams maintaining a recognizable brand voice across all digital channels.
- Uniform voice characteristics per request
- No variation in tone for same preset
- Suitable for branded news segments
Data Privacy and Training
NewsAudio API treats your editorial content as transient data. Input text is used exclusively to synthesize the requested audio file and is never retained for model training or future development. This ensures that your unpublished articles, internal memos, or breaking news drafts remain confidential and do not influence the underlying engine’s behavior.
Input text not used for training
The synthesis process is designed to be stateless regarding your content. Once the audio is generated and delivered, the original text is not stored in any training dataset. This guarantees that your proprietary narratives or sensitive journalistic sources are not ingested into the voice model’s learning pool, maintaining strict separation between your data and the AI’s evolution.
- Text is used only for the immediate synthesis task
- No ingestion into training datasets
- Proprietary content remains confidential
No storage of generated audio
Generated audio files are returned directly in the API response and are not permanently stored on our servers. This minimizes data residency risks and ensures that your audio assets remain under your control. You can download and archive the files immediately after generation without relying on our infrastructure for long-term storage.
- Audio is returned in the response
- No permanent storage on our servers
- Full control over audio archiving
Metadata kept for billing
To ensure accurate billing, we retain request metadata including timestamp, character count, and status. This information is used solely to calculate charges based on the character-based pricing model and to verify that credits were applied correctly. No audio content or source text is included in these records.
- Timestamp, character count, and status retained
- Used only for billing accuracy
- No audio or text in metadata
Formats and Compatibility
The API supports five audio formats: MP3, WAV, Opus, FLAC, and PCM. This flexibility allows newsrooms to choose the right balance between file size and fidelity for their specific workflow, whether distributing via web players or archiving master files.
MP3, WAV, Opus, FLAC, PCM
MP3 offers compressed delivery for quick web streaming. WAV and FLAC provide lossless quality ideal for broadcast masters. Opus delivers efficient compression for modern web players. PCM provides raw, uncompressed data for precise post-production editing.
Browser and player compatibility
MP3 and WAV enjoy universal support across all major browsers and audio players. Opus is widely supported in modern environments. FLAC plays natively in most advanced media software. PCM requires conversion for some legacy web players but offers raw fidelity.
Archival quality options
For long-term storage, FLAC and WAV preserve every detail of the generated speech without lossy compression artifacts. This ensures that master audio files remain pristine for future repurposing or broadcast standards. Use FLAC to maintain perfect audio quality without lossy compression artifacts.

Support and Documentation
NewsAudio API provides straightforward documentation designed for editorial teams who need clear, actionable information. The API docs cover all endpoints, request parameters, and error codes in a structured format. You can quickly find the details needed to integrate the speech generation endpoint into your content management system or publishing workflow.
Support is handled exclusively via email, ensuring that every query is logged and addressed by a dedicated team. There are no phone lines or live chat widgets to navigate, which keeps the support process simple and focused. Responses are provided based on availability, with no SLA guarantees or enterprise contracts required.
Email support only
All support inquiries are directed to a single email address. This approach allows for thorough responses without the pressure of immediate availability. You can send questions about billing, integration issues, or voice selection, and receive a detailed reply in your inbox.
Comprehensive API docs
The documentation covers every aspect of the API, from authentication to response formats. It explains the JSON structure required for the /v1/audio/speech endpoint and details the available voices, languages, and formats. You can also find information on rate limits and character counts to help plan your usage.
- Full endpoint reference with request/response examples
- Authentication and key management guide
- Voice and language parameter details
- Error code explanations
Ready cURL examples
Each documentation page includes copy-paste ready cURL commands. This allows you to test the API directly from your terminal without writing code first. You can quickly verify that your key works and that the voice output meets your quality standards before integrating into your production environment.
Limitations to Consider
While NewsAudio API delivers broadcast-ready audio efficiently, understanding its technical boundaries helps newsrooms plan workflows accurately. The service prioritizes simplicity and speed over granular control, meaning some advanced production features require external processing.
Knowing these constraints early prevents unexpected delays during high-volume publishing days. Teams can adapt their scripts and infrastructure to work within the API’s defined scope without needing complex workarounds.
No SSML support
The API does not support Speech Synthesis Markup Language (SSML). This means you cannot use tags to adjust pauses, emphasis, or pronunciation within a single request. If your scripts require specific pacing for dramatic effect or technical terms, you must handle those adjustments outside the API or pre-segment your text into smaller, carefully crafted requests.
For most news briefs, standard natural phrasing is sufficient. However, if your editorial style relies heavily on precise cadence, you may need to manually break text to control delivery speed between segments.
- No pause tags (<break>)
- No emphasis or pitch control
- No phoneme dictionaries or custom pronunciation lexicons
No voice cloning
NewsAudio uses eight fixed voice presets based on licensed recordings from LibriTTS and LibriVox. You cannot upload a sample of your anchor’s voice to create a custom clone. Each voice has a distinct character, such as the neutral tone of Atlas for news or the warm narrator style of Aria.
This ensures consistency across all generated audio files but limits brand-specific voice identity. If you require a unique voice that matches your station’s specific talent, you would need to select the closest preset or produce audio elsewhere for that specific segment.
No streaming endpoint
The API is designed for batch processing, not real-time streaming. You send a POST request with text, and the API returns a complete audio file once synthesis finishes. There is no WebSocket endpoint for partial audio chunks.
This means latency depends on the length of the text and current queue load. For live broadcast scenarios requiring instant audio generation, you should pre-generate audio files during off-peak hours. The typical speed is roughly real-time, but short phrases may take a few seconds to process.
Use Cases for Newsrooms
Newsrooms require consistent audio production without hiring dedicated voice talent for every update. The NewsAudio API enables editorial teams to convert written scripts into broadcast-ready audio files rapidly, maintaining a uniform brand voice across all channels.
This workflow eliminates the logistical overhead of studio booking and editing. By integrating directly into existing content management systems, teams can automate the conversion of text to speech, ensuring timely delivery of audio content to audiences.
Daily news briefs
Generate concise morning or evening briefs from editorial drafts. Select a neutral voice like Atlas for a professional tone or Luna for a softer, storytelling approach. The API processes standard news text efficiently, delivering clear audio summaries ready for podcast distribution or website embedding.
- Consistent voice for daily episodes
- Quick turnaround from draft to audio
- Support for multiple languages
Article audio versions
Provide audio alternatives for long-form journalism. Split articles into segments if they exceed 4096 characters, then join the resulting audio files seamlessly. This allows readers to consume in-depth reporting hands-free, expanding the reach of your written content to commuters and accessibility users.
- Handling texts up to 4096 characters per request
- Joining split audio files for longer articles
- Support for 23 languages including English, Spanish, and German
Press release audio
Convert press releases into audio format for broader distribution. Use energetic voices like Felix for promotional announcements or Milo for conversational updates. The API ensures the text is rendered clearly, maintaining professional quality for media outlets and stakeholders who prefer listening over reading.
- Clear articulation for official statements
- Multiple voice options for different tones
- Easy integration via OpenAI SDK
Comparison with Other Providers
NewsAudio API distinguishes itself through straightforward, character-based billing without hidden tiers or monthly minimums. Unlike many enterprise-focused competitors, there are no seat limits or complex tiered structures, making cost prediction simple for editorial teams managing fluctuating daily content volumes.
The API supports 23 languages with robust voice coverage, whereas many alternatives restrict native language options or charge premiums for less common tongues. This broad support allows global newsrooms to maintain consistent audio quality across diverse regional broadcasts without negotiating separate contracts for each language.
Pricing transparency
Billing is strictly per character at $0.01 per 1,000 characters, with failed requests refunded in full. There are no monthly subscriptions, no minimum spend requirements, and no per-seat fees. This model ensures that editorial budgets scale linearly with actual usage, avoiding waste on unused capacity.
Top-ups via USDT (TRC-20) provide immediate credit, eliminating invoice delays or VAT complications. The absence of enterprise contracts means teams can start small and scale up instantly without renegotiating terms as their audio output grows.
- $0.01 per 1,000 characters; no monthly minimums
- Failed requests are free
- No per-seat or enterprise contract fees
Language coverage
The API supports 23 languages, including Arabic, Chinese, Hindi, Japanese, and Swahili, among others. All eight voices can speak any supported language, ensuring consistent voice quality across different regions. Auto-detection works for Latin script, but explicit language codes are recommended for non-English languages to ensure accuracy.
Many competitors offer fewer native language options or require premium tiers for broader language support. NewsAudio provides comprehensive coverage suitable for international newsrooms without requiring separate integrations for each language.
- 23 supported languages
- All voices support all languages
- Explicit language setting recommended for non-English
Integration flexibility
The API endpoint matches the OpenAI speech endpoint structure, allowing direct use of official OpenAI SDKs by simply changing the base_url. This eliminates the need for learning a new SDK or writing custom integration code. The JSON request body is identical, making migration from other providers seamless.
Other providers often require custom SDKs or complex authentication flows. NewsAudio’s compatibility with standard tools reduces engineering overhead, allowing technical teams to integrate TTS into existing publishing workflows quickly using familiar HTTP requests.
- OpenAI SDK compatible
- Same JSON structure
- No custom SDK required
Final Recommendations
NewsAudio API offers a practical, transparent solution for newsrooms needing consistent audio output without enterprise overhead. By focusing on character-based pricing and OpenAI SDK compatibility, editorial teams can integrate TTS into their existing workflows with minimal friction. The absence of monthly minimums ensures you only pay for what you generate, while failed requests remain free, reducing risk during initial adoption.
The API’s 23-language support and eight distinct voice presets allow for flexible content creation, from neutral news delivery to energetic promo reads. With input text not used for training and generated audio returned without storage, newsrooms can maintain editorial control and privacy. This section outlines how to leverage these features effectively for daily news briefs, article audio versions, and press releases.
Start with the free trial
Sign up with Google to receive $0.50 in trial credit, valid for 7 days. This allows you to generate approximately 50,000 characters of audio at no cost. Use this period to evaluate voice quality and API responsiveness without financial commitment.
Note that email/password accounts do not include trial credit. If you prefer this method, consider a small top-up to test the system. The trial credit provides a realistic sample of the service’s capabilities before scaling.
- Sign up with Google for $0.50 trial credit
- Credit valid for 7 days
- Generates ~50,000 characters of audio
- Email/password accounts start without trial credit
Test with your actual scripts
Use your real news scripts to assess voice suitability. The API supports 23 languages, with auto-detection for Latin script defaulting to English. For non-English Latin text, explicitly set the language parameter to ensure correct pronunciation.
Test different voices like 'atlas' for neutral news or 'felix' for energetic promos. Verify that the output meets your editorial standards for clarity and tone. The API’s consistent voice quality ensures uniformity across episodes.
- Test voices like 'atlas' (neutral) or 'felix' (energetic)
- Explicitly set language for non-English Latin text
- Verify output for clarity and tone
- Ensure consistent voice quality across episodes
Scale based on volume needs
Pricing is $0.01 per 1,000 characters, with no monthly minimums. This model allows you to scale usage based on your newsroom’s output without fixed overhead. Failed requests are free, protecting your budget during high-volume testing.
For larger volumes, consider top-ups via USDT on the TRON network. The API handles up to 60 requests per minute, processing them sequentially. Under load, requests may queue, but output quality remains consistent.
- Pricing: $0.01 per 1,000 characters
- No monthly minimums or subscriptions
- Failed requests are free
- Top-up via USDT on TRON network
- Rate limit: 60 requests per minute
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.
