Frequently asked questions
Which voice is best for French ads?
All eight voices support French. Aria provides a warm, clear narrator style, while Nova offers a brighter tone and Luna a softer approach. Although the system guesses the language from the script, explicitly setting the language parameter ensures accurate pronunciation for your French content. Any voice can deliver high-quality results for your advertising needs.
Are all voices the same price?
Yes. All voices use the same underlying engine and pricing model. You pay $0.01 per 1,000 characters regardless of the voice selected. Whether you choose Aria, Felix, or Orion, the cost per request remains identical, allowing you to switch voices freely without affecting your advertising budget or billing structure.
Can I clone a voice for my brand?
No. AuraCast does not offer custom voice cloning or the ability to upload your own voice. You must choose from the eight preset voices: Aria, Nova, Luna, Sage, Felix, Milo, Atlas, and Orion. Each preset is based on licensed recordings from LibriTTS-R and LibriVox public-domain audiobook readings.
Do the voices sound robotic?
The voices are based on licensed recordings and natural speech patterns, though every file carries an inaudible watermark marking it as synthetic speech. They are optimized for commercial use and radio spots. While they may lack the subtle nuance of human speech, they are designed to be clear and professional for advertising.
Which voice is best for news?
Atlas is described as neutral and news-oriented, making it a suitable choice for news-style content. Orion is deep and documentary-style, which may also fit news formats. You can test both voices with your specific text to determine which best matches the tone of your radio spot or news update.
Can I use Felix for a promo?
Yes. Felix is described as energetic and promo-focused, making it ideal for promotional content. You can use the API to generate audio files with Felix for your advertising campaigns. The voice supports multiple languages, so you can produce localized promos efficiently using the same energetic tone across different markets.
Is Aria suitable for storytelling?
Aria is described as a warm, clear narrator. Luna is described as soft and storytelling-oriented. While Aria can narrate stories, Luna may be better suited for soft, narrative content. You can test both voices with your script to determine which best fits the specific tone of your storytelling project.
Do I need different voices for different languages?
No. Any voice can speak any of the 23 supported languages. You do not need to assign specific voices to specific languages. However, you should set the language parameter explicitly for non-English Latin-script text to ensure accurate pronunciation, as the system otherwise defaults to English for Latin script.
How do I choose between Atlas and Orion?
Atlas is neutral and news-oriented, while Orion is deep and documentary-style. Choose Atlas for standard news delivery or Orion for a more authoritative, documentary tone. Both voices support all 23 languages. You can generate sample audio for both using the API to compare their performance with your specific script.
Can I change voices mid-campaign?
Yes. You can switch voices between API requests. Each request is independent, so you can use different voices for different spots within the same campaign. There is no limit on the number of voices you can use, and the price remains the same regardless of the voice selected for each request.
Are the voices licensed for commercial use?
The voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain audiobook readings. The generated audio files are returned in the response and not stored by AuraCast. You can use the synthetic speech for commercial purposes, such as radio spots and advertising, without additional licensing fees.
Is there a difference in quality between voices?
All voices use the same engine and pricing model. The difference lies in the preset style: Aria is warm, Nova is bright, Luna is soft, Sage is calm, Felix is energetic, Milo is friendly, Atlas is neutral, and Orion is deep. Quality is consistent across all voices, with variations in tone and character.
Can I use the voices for YouTube ads?
Yes. You can generate audio files in formats like mp3, wav, opus, flac, or pcm. These files can be used in YouTube ads. The voices are synthetic but based on licensed recordings. You can produce high-quality audio content for your video ads efficiently using the AuraCast API.
How do I test voices before buying?
You can sign in with Google to receive a $0.50 free trial credit, valid for 7 days. This allows you to generate approximately 50,000 characters of audio. You can test all eight voices and different languages to evaluate quality before committing to a top-up. Failed requests are also free.
Female Voices
AuraCast offers three distinct female presets optimized for commercial radio and localized advertising. Each voice is derived from licensed LibriTTS-R recordings, ensuring consistent quality for startup campaigns and agency spots without enterprise overhead.
Aria: Warm narrator
Aria provides a clear, warm tone ideal for narrative-driven radio spots. Her delivery remains steady across long texts, making her suitable for brand storytelling where clarity is paramount.
- Best for brand narratives
- Clear articulation
- CC BY 4.0 licensed source
Nova: Bright upbeat
Nova delivers a bright, energetic performance perfect for promotional content. Her upbeat cadence captures attention quickly, making her effective for short-form advertising and dynamic radio spots.
- High energy delivery
- Promotional content
- Attention-grabbing tone
Luna: Soft storytelling
Luna offers a soft, gentle voice suitable for soothing narratives. Her calm delivery works well for bedtime stories or gentle product introductions where a less aggressive tone is preferred.
- Soft delivery
- Storytelling focus
- Gentle tone
Male Voices
Male voices provide distinct tonal options for advertising and radio spot production. Each preset is engineered for specific broadcast contexts, ensuring consistent audio quality across varied campaign requirements without requiring additional engineering overhead.
Felix: Energetic promo
Felix delivers an energetic tone ideal for promotional content and high-impact radio spots. His delivery style supports dynamic pacing, making him suitable for time-sensitive advertisements where attention retention is critical.
- Optimized for short-form promotional copy
- High energy delivery for attention retention
- Ideal for retail and service announcements
Milo: Friendly conversational
Milo provides a warm, approachable voice suitable for conversational content. His natural cadence works well for brand storytelling, podcast intros, and direct-to-consumer messaging where a personal touch enhances listener trust.
- Natural conversational cadence
- Warm tone for brand storytelling
- Suitable for direct-to-consumer messaging
Atlas: Neutral news
Atlas offers a neutral delivery style optimized for news-style readings and informational content. His consistent tone ensures clarity and professionalism, making him a reliable choice for corporate updates and factual reporting.
- Neutral tone for clear communication
- Optimized for news-style readings
- Professional delivery for corporate updates
Specialty Voices
AuraCast offers distinct voice presets optimized for specific broadcast formats, allowing agencies to match tonal requirements without custom production. Each voice is generated from licensed LibriTTS-R recordings, ensuring consistent commercial rights under CC BY 4.0.
The following section details Orion and Sage, two specialized presets designed for documentary and instructional contexts, along with practical usage scenarios for radio advertising.
Orion: Deep documentary
Orion provides a deep, authoritative tone suitable for documentary-style radio spots. Its vocal range conveys gravity and trust, making it ideal for financial services, luxury goods, or serious narrative-driven advertisements.
Use Orion when the brand message requires a commanding presence without sounding aggressive. The voice maintains clarity even at lower speeds, ensuring complex product details are understood.
- Best for: Financial reports, luxury brand storytelling, serious narrative ads
- Tone: Authoritative, deep, trustworthy
- Speed range: 0.25–4.0x
Sage: Calm instructional
Sage delivers a calm, instructional tone perfect for explaining features or guiding listeners through a process. Its clear articulation reduces cognitive load, making it ideal for tech startups or educational content.
Deploy Sage for voiceovers that require patience and clarity, such as app tutorials, service explanations, or step-by-step radio instructions. The voice remains consistent across long texts.
- Best for: Tech explainers, service tutorials, educational spots
- Tone: Calm, clear, patient
- Speed range: 0.25–4.0x
Usage scenarios
For a startup launching a new app, Sage can efficiently explain features in a 30-second spot. Orion suits a premium coffee brand’s radio ad, emphasizing quality and heritage.
Agencies can rotate these voices to test audience response. All eight voices support all 23 languages, enabling localized campaigns without hiring native voice talent for each market.
- Rotate voices for A/B testing in radio campaigns
- Use Orion for brand heritage, Sage for feature explanation
- Supports 23 languages for global reach
Voice Selection Criteria
Selecting the right voice requires aligning vocal characteristics with brand identity. A warm narrator suits lifestyle brands, while an energetic tone fits promotional spots. Consistency in tone builds listener recognition across campaigns.
Tone matching brand
Match the voice’s inherent character to your brand’s core message. Use Aria for approachable, clear storytelling or Felix for high-energy promotions. Ensure the vocal delivery reinforces your brand’s personality without distracting from the message.
- Aria: Warm, clear narrator for lifestyle brands
- Felix: Energetic, promo-focused for sales spots
- Atlas: Neutral, news-style for corporate updates
Audience demographic
Consider the age and preferences of your target listeners. Younger audiences may respond better to bright, upbeat tones like Nova, while older demographics might prefer the calm, instructional clarity of Sage or the deep authority of Orion.
- Nova: Bright, upbeat for younger demographics
- Sage: Calm, instructional for professional audiences
- Orion: Deep, documentary for mature listeners
Medium suitability
Adapt voice selection to the delivery medium. Radio spots benefit from clear, distinct voices like Milo or Luna, ensuring clarity over airwaves. Podcast integrations may require more nuanced tones like Sage or Aria for sustained engagement.
- Milo: Friendly, conversational for radio spots
- Luna: Soft, storytelling for podcasts
- Ensure clarity over airwaves with distinct voices
Technical Specs
The AuraCast API enforces a strict 4,096-character limit per request, counting spaces and punctuation. For longer radio scripts, split the text into segments and join the resulting audio files to maintain high synthesis quality without exceeding payload constraints.
Character limit per voice
Each API call processes up to 4,096 characters. Exceeding this limit requires manual segmentation. All eight voices adhere to this uniform constraint, ensuring consistent processing behavior regardless of voice selection.
- Limit: 4,096 characters per request
- Includes spaces and punctuation
- Manual splitting required for longer scripts
Speed range support
Adjust playback speed between 0.25x and 4.0x to fit precise ad timings. This flexibility allows agencies to align voiceover duration with commercial spot constraints without re-recording.
- Range: 0.25x to 4.0x
- Adjustable per request
- Optimizes ad spot duration
Format compatibility
Output supports MP3, WAV, OPUS, FLAC, and PCM formats. Select the appropriate codec based on your broadcasting infrastructure requirements, balancing file size against audio fidelity for radio delivery.
- MP3, WAV, OPUS, FLAC, PCM
- Configurable via response_format
- Suitable for radio broadcasting
Licensing
AuraCast’s synthetic voices are generated from licensed recordings in the LibriTTS-R dataset, sourced from LibriVox public-domain audiobook readings. This foundation ensures that the acoustic characteristics of our eight presets—Aria, Nova, Luna, Sage, Felix, Milo, Atlas, and Orion—are derived from high-quality, legally cleared human speech.
The underlying data operates under a CC BY 4.0 basis, providing a transparent lineage for all generated audio. The voices are derived from this licensed material, ensuring that the resulting radio spots and localized content carry a clear, traceable origin without ambiguous rights issues.
LibriTTS-R source
Our voice models are trained on the LibriTTS-R dataset, which contains hours of read audiobooks. This source provides the diverse phonetic and prosodic variety necessary for natural-sounding synthetic speech across 23 languages, ensuring that every generated audio file maintains high fidelity suitable for commercial broadcast.
CC BY 4.0 basis
The CC BY 4.0 license governs the underlying speech data, allowing for broad usage rights while requiring appropriate attribution. This legal framework supports our infrastructure’s reliability, ensuring that the synthetic voices you generate for radio ads remain clear of copyright disputes that often plague custom voice recordings.
Commercial usage rights
All generated audio files are cleared for commercial use in advertising, radio spots, and localized content. You do not need to negotiate additional licenses for the voice talent, as the synthetic output is distinct from the original recordings. This streamlines your production workflow, allowing immediate deployment of audio assets.
- No additional licensing fees per spot
- Cleared for broadcast and digital media
- Attribution requirements depend on the specific license of the source recordings
Pricing Impact
AuraCast charges a flat rate of $10 per 1 million characters, regardless of which of the eight voices you select. Whether you choose the warm narration of Aria or the deep documentary tone of Orion, the cost per character remains identical. This uniformity eliminates hidden costs for premium voice tiers.
There are no voice-specific fees or tiered pricing structures. Your budget depends solely on character volume and speed adjustments, not voice personality. This predictability allows advertising agencies to forecast production costs accurately for large-scale radio spot campaigns without worrying about variable voice rates.
Same price for all voices
All eight preset voices share the same base rate. You do not pay extra for high-fidelity models like tts-1-hd versus standard tts-1 if using the same engine tier. The pricing model is strictly volume-based, ensuring consistent unit economics across all creative assets.
This uniform pricing applies to all response formats including mp3, wav, opus, flac, and pcm. Whether you need compressed audio for digital delivery or lossless files for broadcast, the character count dictates the bill, not the file type.
- Flat $0.01 per 1,000 characters across all voices
- No premium tier for specific voice personalities
- Identical rates for tts-1 and tts-1-hd models
No voice-specific fees
You are not charged extra for switching voices between requests. Each API call incurs the same per-character cost, allowing you to A/B test different voices for the same ad copy without incurring differential pricing. This flexibility supports dynamic ad insertion strategies.
Failed requests are free, ensuring you only pay for successfully generated audio. This protects your budget during testing phases where voice suitability is being evaluated against specific demographic targets.
Volume efficiency
Longer texts are split into 4096-character chunks and joined by the client, ensuring efficient processing without penalty. Typical generation speed is roughly real-time, with short phrases taking a few seconds. Requests are processed sequentially, so under load, a request may wait in a queue.
With a minimum top-up of $5 via USDT (TRC-20), small startups can start immediately. Balances do not expire, allowing you to accumulate credit for high-volume campaign launches without monthly subscription constraints.

Quality Comparison
AuraCast voices are generated via a cloud API, ensuring uniform audio characteristics across all production calls. This consistency eliminates the variability inherent in human recording sessions, providing advertising agencies with predictable acoustic profiles for repeated brand messaging.
The synthetic nature of the output is marked by an inaudible watermark, distinguishing generated audio from human recordings without affecting listener perception. This marker aids in content verification while maintaining the professional quality expected in radio spots and localized advertising campaigns.
Synthetic watermark
Every audio file generated through the API carries an inaudible watermark identifying it as synthetic speech. This feature allows for easy verification of content origin without altering the listening experience for the end consumer.
- Inaudible to human ears
- Applied to all formats (mp3, wav, opus, flac, pcm)
- No impact on audio quality or latency
Consistency across calls
Unlike human voice actors who may vary in performance due to fatigue or environment, our infrastructure delivers consistent synthetic voices. This ensures that your radio spots maintain the same tonal quality and pacing across all episodes and campaigns.
- No variability due to recording conditions
- Uniform tone across 8 presets
- Ideal for serialized advertising content
Noise level
Generated audio is clean and free from background studio noise or breath sounds unless specifically simulated by the engine. This results in a crisp delivery suitable for direct broadcast or further post-production mixing, reducing the need for extensive audio cleanup.
- Clean signal output
- Minimal background artifacts
- Ready for immediate broadcast or mixing
Integration Tips
AuraCast API mirrors the OpenAI speech endpoint structure, allowing your existing integration code to function with a simple base_url adjustment to https://api.km12.info/v1. Authentication requires a Bearer token in the header, ensuring secure access to your generated audio assets for broadcast.
SDK configuration
Utilize standard OpenAI SDKs by pointing them to the AuraCast base URL. This allows your engineering team to leverage familiar libraries without rewriting core logic. Ensure your Authorization: Bearer header contains the active key issued from your /get-api-key/ dashboard.
- Set base_url to https://api.km12.info/v1
- Use standard JSON body structure
- Include Bearer token in headers
Batch voice selection
Generate multiple radio spots by iterating through the 8 available voices in a single script. Since all voices share the same pricing and technical limits, you can efficiently test Aria for warmth or Felix for energy without incurring variable costs.
- Test all 8 presets simultaneously
- Maintain consistent pricing structure
- Automate voice rotation via API
Dynamic voice switching
Switch voices programmatically based on campaign metadata. Pass the desired voice ID (e.g., 'nova' or 'atlas') in the request body to match specific ad segments. This enables automated localization and tone adjustments across different regional markets.
- Pass voice ID in JSON body
- Match tone to campaign goals
- Automate segment-specific voice assignment
Common Mistakes
Using the wrong voice for your target audience is a frequent error in radio spot production. Pairing a deep, documentary-style voice with a high-energy promotional script creates cognitive dissonance, reducing listener retention and brand recall.
Mismatched tone
Ensure the voice character aligns with the script's emotional intent. For upbeat ads, select Nova or Felix; for serious news-style updates, use Atlas or Orion. Consistent tone reinforces the message.
Ignoring language codes
If your text is in Latin script but not English, explicitly set the language parameter. Without it, the system defaults to English, potentially mispronouncing non-English words or using incorrect intonation patterns.
Over-speeding
Increasing speed beyond 1.0x can make synthetic speech sound unnatural or rushed, especially for complex product names. Stick to the 0.25–4.0 range, but test readability at higher speeds to maintain clarity for your listeners.
Future Proofing
AuraCast’s infrastructure processes requests sequentially, ensuring consistent audio delivery even during peak advertising campaign periods. The system processes requests sequentially, maintaining reliability without requiring complex load-balancing strategies from the client side.
Standardized API design ensures that your integration remains compatible with future updates. By adhering to established patterns, your application can adapt seamlessly as new capabilities are introduced, minimizing maintenance overhead for your engineering team.
Scalable infrastructure
The backend processes requests sequentially, ensuring stable performance under variable loads. This design requires client-side queue management under load, allowing your automated ad delivery systems to operate reliably during high-volume periods.
API versioning
We maintain backward compatibility within the v1 endpoint structure. Your existing integration code will continue to function as we introduce new voice presets or format options, reducing the risk of service disruption during updates.
Standard compliance
The API follows a widely adopted JSON structure, compatible with standard HTTP clients. This ensures easy integration with existing dev tools and libraries, supporting long-term technical viability without vendor lock-in.
Conclusion
AuraCast delivers eight consistent synthetic voices optimized for commercial radio spots and localized advertising content. Each voice provides distinct tonal qualities, from warm narration to energetic promotion, ensuring your brand message resonates with specific audience demographics without the overhead of enterprise contracts.
Summary of best uses
Use Aria or Luna for soft storytelling segments, while Felix suits energetic promos. Atlas and Orion provide neutral or deep tones for news and documentary styles. All voices support 23 languages, enabling efficient localization of ad campaigns across global markets with consistent audio quality.
- Aria and Luna for narrative ads
- Felix for high-energy promotions
- Atlas and Orion for authoritative news tones
Final recommendations
Select voices based on brand tone and target demographic. Always specify the language code for non-English text to ensure accurate pronunciation. Stick to the 4096-character limit per request to maintain processing efficiency and avoid queue delays during peak usage periods.
- Match voice tone to brand identity
- Set explicit language codes for Latin-script text
- Keep inputs under 4096 characters
Next steps
Generate your API key via the dashboard to start testing voices immediately. Review the documentation for integration details with the OpenAI SDK. Top up your balance with USDT (TRC-20) to ensure uninterrupted production of your radio spot inventory.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. This ensures that your radio spots maintain the same tonal quality and pacing across all episodes and campaigns.
23 languages
Any voice speaks any of these languages.
