Frequently asked questions
Which voice is best for corporate training?
Q: Which voice is suitable for corporate training? A: Sage offers a calm, instructional tone ideal for training modules, while Atlas provides a neutral, news-style delivery suitable for professional updates. Felix works well for energetic promotional content within corporate learning environments. Select the voice that matches the specific tone of your course material to ensure consistent engagement across your library.
Can any voice speak any language?
Yes, all eight voice presets can speak any of the 23 supported languages. If you do not specify a language code, the system guesses based on the script, defaulting to English for Latin text. For non-English Latin-script text, you should explicitly set the language parameter to ensure accuracy.
Is there a price difference between voices?
No, there is no price difference between voices. All voices use the same engine and pricing structure. You pay $10 per 1 million characters regardless of which voice you select. The cost depends solely on the length of the input text and the response format chosen.
Are the voices realistic?
The voices are based on licensed recordings from LibriTTS-R and LibriVox public-domain audiobook readings. They sound natural but carry an inaudible watermark marking them as synthetic speech. They are not perfect human replicas but provide high-quality, consistent narration suitable for professional educational content without the variability of human recording sessions.
Can I change the speed of the voice?
Yes, you can adjust the speech speed between 0.25 and 4.0. This allows you to slow down complex instructional content or speed up simple reviews. Adjust the speed parameter in your API request to tailor the pacing to your audience’s needs, ensuring clarity and engagement for different types of learning material.
Do the voices sound robotic?
The voices are based on real licensed recordings and do not sound overly robotic. They offer natural intonation and pacing suitable for professional use. However, they are synthetic and carry an inaudible watermark. They provide consistent quality without the emotional nuances of human actors, which is ideal for standardized corporate training.
Which voice is best for kids' courses?
Q: Which voice is suitable for kids' courses? A: Nova offers a bright, upbeat tone suitable for engaging younger audiences, while Luna provides a soft, storytelling voice ideal for narrative content. Aria is another warm option for clear narration. Choose based on whether you need energetic engagement or gentle storytelling to match the specific style of your children’s educational content.
Can I use the voices in video ads?
Yes, you can use the generated audio in video ads. Felix is energetic and well-suited for promotional content. The voices are based on licensed recordings, so you can integrate them into your marketing materials. Ensure your usage complies with the terms of the API service and the underlying license of the source recordings.
Are the voices gender-neutral?
No, the voices are distinctly gendered. There are four female voices (Aria, Nova, Luna, Sage) and four male voices (Felix, Milo, Atlas, Orion). Each has a specific tonal character. You can select the gender that best fits your brand voice or the persona of your course instructor to maintain consistency.
Do I need to credit the voices?
The voices are based on licensed recordings from LibriVox public-domain audiobook readings. While the API service handles the licensing, it is good practice to acknowledge the source. The generated audio carries an inaudible watermark marking it as synthetic. Check the specific license terms for LibriTTS-R to ensure proper attribution if required by your distribution channel.
Can I use the voices offline?
Yes, once you receive the audio file from the API, you can use it offline. The API returns standard audio formats like MP3, WAV, OPUS, FLAC, or PCM. You can download these files and embed them in your LMS, video projects, or mobile apps without needing an active internet connection during playback.
Is there a limit to how many times I can use a voice?
There is no limit to how many times you can use a voice. You can generate audio with any of the eight voices as many times as your balance allows. Each request consumes characters based on the text length. You can switch between voices freely without any additional restrictions or fees.
Female Preset Voices for Courses
Narrata’s female preset voices provide consistent narration for learning management systems, ensuring uniform audio quality across large course libraries. These voices are designed for clarity and engagement, allowing instructional designers to maintain a professional tone without the variability of human recording sessions.
Aria: Warm and Clear Narrator
Aria delivers a warm, clear narrator tone ideal for standard educational content. Her voice ensures that complex instructional material is easy to follow, making her suitable for general knowledge modules where clarity is paramount.
- Optimized for clear instruction delivery
- Warm tone reduces listener fatigue
Nova: Bright and Upbeat
Nova offers a bright, upbeat style that adds energy to course introductions or lighter learning segments. This preset is particularly effective for keeping learners engaged in short modules or promotional training content.
- High energy for engaging content
- Ideal for introductions and short modules
Female Preset Voices for Storytelling
Luna and Sage bring narrative depth to educational content, transforming static text into engaging audio lessons. Luna’s soft, storytelling tone adds warmth to case studies and historical narratives, while Sage’s calm, instructional voice ensures clarity during complex concept explanations.
Luna: Soft Storytelling Tone
Luna delivers a warm, clear narration style ideal for immersive learning modules. Her voice maintains a gentle pace that helps learners retain information without feeling rushed, making her suitable for longer text inputs up to 4,096 characters per request.
- Best for historical narratives and case studies
- Maintains listener engagement over longer durations
- Supports all 23 languages with consistent warmth
Sage: Calm and Instructional
Sage provides a steady, instructional baseline perfect for procedural training and technical guides. Her neutral yet approachable tone reduces cognitive load, allowing learners to focus on the material rather than the delivery style.
- Ideal for step-by-step technical tutorials
- Reduces listener fatigue during dense content
- Works effectively across all voice formats including opus and flac
Male Preset Voices for Promos
Felix and Milo offer distinct male voices optimized for promotional and conversational e-learning content. Felix delivers an energetic, high-impact tone ideal for short marketing clips or module introductions within your LMS. Milo provides a friendly, approachable style that feels like a colleague guiding the learner, perfect for walkthroughs or soft skills training.
These presets ensure consistent audio quality across large course libraries. Both voices are generated instantly via the API, allowing you to maintain a unified brand voice across diverse training materials.
Felix: Energetic Promo Style
Felix is designed to capture attention quickly. His delivery is crisp and dynamic, making him suitable for product highlights, certification announcements, or brief motivational segments in onboarding modules. The energetic tone helps sustain learner engagement during short-form content.
- Ideal for module introductions and promotional clips
- High energy delivery for quick engagement
- Consistent tone across all generated audio
Milo: Friendly and Conversational
Milo sounds like a supportive peer rather than a lecturer. This voice is excellent for step-by-step instructions, soft skills scenarios, or Q&A style content. His conversational style reduces the cognitive load for learners, making complex topics feel more accessible.
Use Milo for narrative-driven training where a personal touch enhances retention. The voice remains steady and clear, ensuring comprehension even in noisy office environments where employees might listen to audio courses.
- Best for step-by-step tutorials
- Approachable, peer-like tone
- Enhances retention in narrative training
Male Preset Voices for News and Docs
For news updates or factual reporting, Atlas provides a neutral, professional tone that ensures clarity without emotional bias. This preset is ideal for daily briefings or factual training modules where the content itself carries the weight, allowing the voice to remain unobtrusive and consistent across large libraries.
Atlas: Neutral News Anchor
Atlas delivers clear, steady narration suitable for compliance training or factual overviews. Its neutral timbre ensures that the message is received without stylistic distraction, making it a reliable choice for corporate communications where consistency is prioritized over personality.
- Best for: factual reporting, compliance updates, and standard corporate training.
Orion: Deep Documentary Style
For deeper narrative contexts, Orion offers a rich, authoritative voice reminiscent of high-quality documentary narration. This preset adds gravitas to historical modules or in-depth technical explanations.
- Best for: documentary-style courses, historical narratives, and in-depth technical explanations.
Choosing the Right Voice for Your Audience
Selecting the appropriate voice ensures your training materials resonate with the intended learners. A mismatch in tone can distract from the educational content, reducing retention and engagement. Aligning vocal characteristics with the subject matter helps maintain a consistent narrative voice across large course libraries.
Matching Tone to Content Type
The content type should dictate the voice selection. Use Felix for energetic promotional modules or Milo for friendly, conversational explanations. For structured compliance or news-style updates, Atlas provides a neutral anchor tone, while Orion suits deep, documentary-style learning. Luna works well for soft storytelling, and Sage offers a calm instructional presence.
Aria’s warm, clear narration is ideal for general course introductions. Nova’s bright, upbeat style keeps learners engaged in dynamic topics. Selecting the right preset ensures the audio matches the instructional intent.
- Felix: Energetic promo style for introductory modules
- Milo: Friendly, conversational tone for explanations
- Atlas: Neutral news anchor style for compliance
- Orion: Deep documentary style for in-depth learning
- Luna: Soft storytelling for narrative content
- Sage: Calm instructional tone for structured lessons
Audience Preferences
Consider the demographic and learning context of your audience. Corporate learners may prefer the neutral clarity of Atlas or the calm instruction of Sage. Younger audiences might engage better with Nova’s upbeat energy or Felix’s promotional drive. Testing different voices with sample modules can help determine which style resonates best with your specific user base.
Consistency is key. Once a voice is selected for a course series, maintaining that choice helps learners familiarize themselves with the narrator’s style, reducing cognitive load and improving the overall learning experience.
Language Support for Global Courses
Narrata supports 23 languages for global e-learning, including English, Spanish, German, French, Japanese, Korean, Chinese, Arabic, Hindi, and Russian. This allows L&D teams to localize training content consistently without managing multiple voice providers or negotiating separate contracts for different regions.
23 Languages Available
The API covers major markets such as Chinese, Danish, Dutch, Finnish, Greek, Hebrew, Italian, Malay, Norwegian, Polish, Portuguese, Swahili, Swedish, and Turkish. It does not support Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, or Bengali. For Latin-script text outside English, explicitly setting the language parameter ensures accurate pronunciation instead of relying on default guessing.
- Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish
Any Voice Speaks Any Language
All eight preset voices can produce any of the 23 supported languages. You do not need to assign specific voices to specific languages; you can use Aria for French modules or Felix for Spanish promos. This simplifies your content library management and ensures a consistent brand tone across all localized training materials.
Technical Specifications
Narrata’s TTS API supports five standard audio formats to ensure compatibility with your Learning Management System. You can generate files in mp3, wav, opus, flac, or pcm formats via the response_format parameter in your JSON request body.
Adjust playback speed between 0.25 and 4.0 times normal rate using the speed field. This allows you to fine-tune listening comfort for complex training modules or quick refresher courses without re-generating the audio file.
Audio Formats Supported
The API returns audio files in mp3, wav, opus, flac, or pcm formats. Select the format that best fits your LMS storage constraints and playback requirements. MP3 offers efficient compression, while WAV and FLAC provide lossless quality for high-fidelity training materials.
Speed Control Options
Set the speed parameter between 0.25 and 4.0 to adjust voice tempo. Slower speeds aid comprehension for dense technical content, while faster speeds suit repetitive or review modules. Requests are processed one after another, so under load a request may wait in a queue.
Note that requests are processed one after another, so under load a request may wait in a queue regardless of speed setting.

Pricing and Usage for Voice Selection
Every voice preset costs the same: $10 per million characters, regardless of whether you select Aria, Orion, or any other option. This flat rate simplifies budgeting for large L&D teams generating diverse course content. You only pay for the text length, including spaces and punctuation, with no hidden fees per voice type.
Failed synthesis requests are free, ensuring you do not pay for errors. The standard rate applies uniformly, making it easy to predict costs for scaling audio production across hundreds of modules without worrying about voice-specific premiums.
Same Price for All Voices
All eight voice presets are priced identically at $0.01 per 1,000 characters. Whether you need the warm narration of Aria for onboarding or the deep documentary style of Orion for compliance training, the cost per character remains constant. This uniform pricing allows you to experiment with different tones for various audience segments without impacting your overall budget.
- $10 per 1 million characters ($0.01 per 1,000 characters)
- All 8 voices share the same base rate
- Spaces and punctuation count toward character limits
No Extra Cost for HD
Choosing the high-definition model (tts-1-hd) does not increase your per-character cost. Both "tts-1" and "tts-1-hd" use the same engine pricing structure. You can opt for higher audio fidelity for executive summaries or detailed technical modules without paying a premium for the HD designation. The only difference is the output quality, not the price.
- tts-1 and tts-1-hd share identical pricing
- No surcharge for higher fidelity
- Choose HD based on quality needs, not cost
Licensing and Attribution
Narrata’s audio generation relies on voice models derived from licensed recordings of LibriTTS-R, distributed under the CC BY 4.0 license. This source material originates from LibriVox public-domain audiobook readings, providing a legally sound foundation for commercial course narration without requiring complex rights clearance for each voice model.
CC BY 4.0 Source Material
Using CC BY 4.0 licensed audio ensures that the underlying voice synthesis models are legally cleared for enterprise use. This reduces legal risk when deploying audio across large educational content libraries, as the base recordings are already cleared for derivative works.
Watermarked Output
Every generated audio file contains an inaudible watermark that identifies the content as synthetic speech. This watermark serves as a clear marker of AI origin, helping maintain transparency with learners and ensuring compliance with institutional policies regarding AI-generated media in training modules.
Integration with LMS Platforms
Integrate the TTS API directly into your Learning Management System by pointing requests to the API base URL. The endpoint accepts standard JSON bodies, making it compatible with most modern LMS platforms that support webhook or plugin-based API calls. This allows your team to generate audio narrations on demand.
By using the same request shape as the OpenAI speech endpoint, you can leverage existing HTTP libraries or SDKs to automate voice generation. Simply provide your API key in the authorization header and send the text content along with your preferred voice preset and format settings.
API Integration Steps
Begin by obtaining your API key from the /get-api-key/ dashboard. Send a POST request to https://api.c168s.info/v1/audio/speech with your text input, selected voice, and desired audio format. The system processes the request and returns the generated audio file, which you can then store or link directly within your LMS modules.
Ensure your integration handles the 4096-character limit per request for longer course segments. You can split longer scripts into multiple requests and concatenate the resulting audio files to maintain seamless narration across entire lessons or training modules.
- Use POST requests to the /v1/audio/speech endpoint
- Include Bearer token in Authorization header
- Specify voice, format, and speed in JSON body
- Handle 4096 character input limit per request
Automation Workflows
Automate voice generation by connecting your LMS content management system to the API. When new course content is published, trigger an API call to generate the audio narration automatically. This eliminates manual recording steps and ensures consistent audio quality across all your training materials.
You can schedule bulk audio generation for entire course libraries during off-peak hours. By pre-generating audio files, you reduce latency for end-users and ensure that all learners access high-quality narration immediately upon enrolling in your modules.
- Trigger API calls on content publish
- Pre-generate audio for bulk libraries
- Schedule tasks during off-peak hours
- Automate voice selection based on content type
Common Use Cases in Education
TTS APIs enable consistent audio generation across thousands of hours of content, ensuring uniform quality for large educational libraries. This standardization is critical for maintaining brand voice across diverse training modules without the variability of multiple voice actors.
By automating narration, L&D teams can rapidly update or localize onboarding materials. The API’s character-based pricing allows predictable budgeting for high-volume course production, making it feasible to generate audio for entire curriculums at scale.
Onboarding Modules
New hire onboarding often requires standardized audio across multiple departments. Using preset voices like Aria or Felix ensures a consistent tone for welcome videos and policy overviews. Teams can generate audio for hundreds of employee profiles, reducing production time.
The ability to quickly regenerate audio with updated text means onboarding materials stay current. When company policies change, L&D teams can update the source text and regenerate the audio without re-recording, ensuring all new hires receive the latest information.
- Generate audio for welcome videos in minutes.
- Update policy content without re-recording sessions.
- Maintain consistent tone across departments.
Compliance Training
Compliance modules require precise, clear narration to ensure regulatory details are understood. Voices like Sage or Atlas provide a calm, instructional tone suitable for explaining complex regulations. The API’s 4096-character limit per request allows for concise, focused segments that improve learner retention.
Since failed requests are free, teams can test audio quality for compliance content without financial risk. The 60 requests per minute limit supports batch processing, allowing compliance officers to generate audio for entire training decks during off-peak hours.
- Clear narration for complex regulations.
- Free retries for quality assurance.
- Batch processing for large training decks.
Limitations to Consider
While the platform offers eight distinct preset voices, it does not support custom voice cloning. You cannot upload a sample recording to create a unique voice model, nor can you design a voice from scratch. This means every generated audio file uses one of the eight licensed presets derived from LibriTTS-R and LibriVox sources.
Additionally, the API does not support SSML (Speech Synthesis Markup Language). You cannot use SSML tags to control precise pauses, emphasis, or pronunciation within your text. This limits granular audio control to the voice preset selection, speed adjustment, and basic text input, making it suitable for standard course narration but less ideal for complex audio engineering needs.
No Custom Cloning
The service provides eight fixed preset voices rather than allowing custom voice cloning. You cannot upload a sample recording to create a unique voice model, nor can you design a voice from scratch. Every generated audio file uses one of these licensed presets, which are based on recordings from LibriTTS-R and LibriVox public-domain audiobook readings.
This approach ensures consistency and legal clarity for enterprise L&D teams, as the voices are derived from licensed material. However, it means you cannot replicate a specific trainer’s voice or create a branded audio persona unique to your organization without using the existing presets.
- No custom voice cloning available
- No uploading your own voice samples
- No voice design tools
- Fixed eight preset voices only
No SSML Support
The API does not support SSML (Speech Synthesis Markup Language), so you cannot use tags to control precise pauses, emphasis, or pronunciation within your text. This limits granular audio control to the voice preset selection, speed adjustment, and basic text input.
For most educational content, this is sufficient. However, if your course requires specific phonetic pronunciations or complex audio engineering, you may need to pre-edit your text or use external tools to manage pacing before sending it to the API.
- No SSML tag support
- No precise pause control
- No emphasis tags
- No pronunciation dictionaries
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. This preset is particularly effective for keeping learners engaged in short modules or promotional training content.
23 languages
Any voice speaks any of these languages.
