Choosing the Right Voice for Your Online Courses

By CourseVoice editorial team · Updated 2026-11-19

This guide helps startup teams building online courses and LMS platforms select the right text-to-speech voice for their lesson scripts. We break down which of our eight presets fits specific content types, from warm narrators to deep documentary voices, ensuring your audio matches your pedagogical tone.

You will learn how to handle language support, manage character limits, and integrate with the OpenAI SDK. We cover pricing models, technical constraints, and common pitfalls so you can generate professional voice-overs efficiently without unexpected costs or integration hurdles.

Choosing the Right Voice for Your Online Courses

Frequently asked questions

What are the best voices for online courses?

Q: Which voices are suitable for online courses? A: Sage offers a calm, instructional tone ideal for lessons. Luna provides a soft, storytelling style for narrative content. Felix suits energetic promo segments, while Atlas delivers a neutral, news-like delivery for clear information. Choose based on your script’s mood, as all eight presets support every language.

Which languages are supported by the API?

The API supports 23 languages: 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. Set the language parameter for non-English Latin text to ensure accuracy.

How does the pricing compare to human voice-overs?

Pricing is $0.01 per 1,000 characters. This is significantly lower than hiring human voice actors per script. Failed requests are free, ensuring you only pay for successful audio generation. There are no monthly plans or minimum spend requirements, making it cost-effective for variable course production volumes.

Is the API easy to integrate for non-developers?

The endpoint matches the OpenAI speech endpoint shape. You can use the official OpenAI SDK by changing the base URL to https://api.api36.info/v1. The /get-api-key/ page provides a ready curl example. This standard format simplifies integration for developers familiar with common API patterns.

What happens if I exceed the rate limit?

The limit is 60 requests per minute per key (10 before the first top-up). Requests are processed sequentially, so excess requests may wait in a queue. You can issue a new key at any time to reset limits, though the old key stops immediately. Monitor usage via the dashboard.

Can I use the API for commercial courses?

Yes. The API generates audio files that you can use in your online courses. Voices are based on licensed recordings from LibriTTS-R and LibriVox. Each file carries an inaudible watermark marking it as synthetic speech. There are no restrictions on commercial usage of the generated audio files.

How long does it take to generate audio?

Typical speed is roughly real-time, about 25 characters per second of generation. Short phrases take a few seconds. Under load, requests may wait in a queue since they are processed one after another. Longer texts take longer proportional to their character count.

Is the audio quality consistent across languages?

Quality is consistent as all eight voices support all 23 languages. The engine processes each language uniformly. However, ensure you set the language parameter for non-English Latin text to avoid default English guessing. This ensures accurate pronunciation and tone across different linguistic contexts.

Do I need to pay for every request?

You pay for successful requests. Failed requests are free, and the charge is returned in full if synthesis fails. You are charged $0.01 per 1,000 characters. Spaces and punctuation count as characters. Failed attempts do not incur any cost, protecting your budget.

Can I store the generated audio myself?

Yes. The API returns the audio file in the response. It is not stored by the service after delivery. You are responsible for storing the generated audio files on your own infrastructure. Request metadata is kept for billing, but the audio content is yours to manage.

What is the difference between tts-1 and tts-1-hd?

Both models use the same engine and have the same price. The only difference is the model identifier in the request. You can choose either 'tts-1' or 'tts-1-hd' without any change in cost or performance characteristics. They produce audio with the same engine and price.

Are there any hidden fees?

No hidden fees exist. Pricing is strictly $0.01 per 1,000 characters. Top-ups via USDT (TRC-20) are credited automatically. There are no subscriptions, seats, or minimum spends. Failed requests are free. The balance does not expire. All costs are transparent and visible in your account dashboard.

How do I handle long course scripts?

Split texts into parts of up to 4096 characters per request. Join the resulting audio files to create the final track. The API processes each part sequentially. Ensure your implementation handles the concatenation of multiple audio files generated from separate requests to maintain continuity.

Is the API reliable for high traffic?

Requests are processed one after another, so under load, a request may wait in a queue. There is no SLA or uptime guarantee. The rate limit is 60 requests per minute per key. For higher traffic, consider issuing new keys or managing request queues in your application.

Criteria for Course Content TTS

Sound wave illustration for: Criteria for Course Content TTS

Selecting the right voice is critical for online course retention. Our API offers eight distinct presets, from warm narrators like Aria to instructional tones like Sage, ensuring your lesson scripts sound professional and consistent.

Language support covers 23 languages, with automatic detection for Latin scripts defaulting to English. Explicit language codes prevent mispronunciation in mixed-language content, ensuring clarity for global students.

Voice Quality and Tone

Choose voices that match your pedagogy. Use Luna for storytelling modules or Felix for energetic promo segments. Each voice is based on licensed recordings, providing clear, synthetic speech without celebrity imitation.

  • Aria: warm, clear narrator
  • Sage: calm, instructional
  • Felix: energetic, promo
  • Luna: soft, storytelling

Language Support

Support 23 languages including English, Spanish, German, and Japanese. Any voice can speak any supported language. For non-English Latin text, always set the language parameter to avoid default English guessing.

  • Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish

Ease of Integration

Integrate via standard HTTP or OpenAI SDKs by changing the base_url. Send a POST request to /v1/audio/speech with your text and desired voice. Keys are issued instantly upon signup.

Voice Options for Different Content Types

Sound wave illustration for: Voice Options for Different Content Types

Selecting the right voice persona ensures your course audio matches the learning objective. EduEcho provides eight distinct presets, each optimized for specific content styles. This allows you to maintain a consistent auditory identity across different modules without hiring multiple voice actors.

The following options cover the primary formats used in online education, from structured lessons to deep-dive documentaries.

Narrator Voices for Lessons

For structured lessons, use aria or sage. Aria offers a warm, clear narration style ideal for reading textbook material. Sage provides a calm, instructional tone that helps students focus on key concepts without distraction.

These voices are best for core curriculum content where clarity and steady pacing are essential for retention.

  • Use aria for general reading passages.
  • Choose sage for instructional or step-by-step guides.

Conversational Voices for Tutorials

Tutorials benefit from a friendly, engaging tone. Nova and milo bring a bright, upbeat energy that keeps learners engaged during software demos or practical exercises. Their conversational style mimics a human instructor guiding you through a process.

This approach reduces the monotony often associated with technical instruction, making complex steps feel more approachable.

  • Select nova for upbeat, energetic tutorials.
  • Pick milo for friendly, conversational explanations.

Documentary Voices for Deep Dives

For in-depth analysis or case studies, atlas and orion provide authoritative depth. Atlas delivers a neutral, news-like delivery suitable for factual reporting. Orion offers a deep, documentary style that adds gravitas to long-form content.

These voices are designed to sustain attention during longer, information-dense segments where a serious tone is appropriate.

  • Use atlas for neutral, factual content.
  • Choose orion for deep, documentary-style narratives.

Language Coverage for Global Reach

Sound wave illustration for: Language Coverage for Global Reach

Reaching a global audience requires clear audio in multiple languages. EduEcho supports 23 languages, including English, Spanish, German, French, Japanese, and Hindi. This allows startups to localize course content without hiring separate voice actors for each region.

The API ensures accurate pronunciation by guessing the language from the text script. For non-English Latin-script text, explicitly setting the language code prevents misinterpretation. Note that Vietnamese, Indonesian, and Thai are not supported.

23 Supported Languages

We support 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 flexibility in tone regardless of the target language.

  • Arabic, Chinese, Danish, Dutch, English, Finnish
  • French, German, Greek, Hebrew, Hindi, Italian
  • Japanese, Korean, Malay, Norwegian, Polish, Portuguese
  • Russian, Spanish, Swahili, Swedish, Turkish

Automatic Language Detection

If you do not set the language parameter, the API guesses the language from the script. Latin script text defaults to English. This simplifies integration for monolingual courses but requires caution for mixed-language content.

Explicit Language Setting

For non-English Latin-script text, always set the language parameter. This ensures the correct voice accent and pronunciation. Without this setting, the API may default to English, resulting in unnatural audio for learners.

Pricing Models Compared

Sound wave illustration for: Pricing Models Compared

EduEcho uses a straightforward pay-per-character model, eliminating complex subscription tiers. At $10 per million characters, you pay only for what you generate. Failed requests are free, ensuring you never pay for errors. This model suits startups with variable content production needs, as there are no minimum spends or monthly commitments.

Pay Per Character vs. Subscription

Our pricing is simple: $0.01 per 1,000 characters, including spaces and punctuation. There are no monthly plans, seats, or subscriptions. This flexibility allows you to scale costs directly with your course output, avoiding fixed fees during quieter periods.

  • Only pay for successful synthesis
  • No monthly commitments or seats
  • Balance never expires

Cost Per Minute of Audio

Audio length correlates with input length. Long texts generate at roughly real-time speed (25 characters per second). Short phrases take a few seconds to process. While file duration varies by voice and speed, your cost is determined strictly by the character count of your input text, not the audio length.

Trial and Top-Up Options

Sign in with Google to receive a $0.50 trial credit, valid for 7 days (approx. 50,000 characters). Email-based accounts start without trial credit. Top-ups are available via USDT (TRC-20) with a $5 minimum. Credits are automatic and do not expire.

Note: We do not accept credit cards or PayPal. All payments are processed via blockchain transfers.

  • Google sign-in: $0.50 free trial
  • USDT (TRC-20) top-ups only
  • Minimum top-up: $5

Technical Integration Options

Sound wave illustration for: Technical Integration Options

Integrating EduEcho into your LMS is straightforward because the API mirrors the OpenAI speech endpoint structure. You can use existing OpenAI SDKs by simply updating the base_url to https://api.api36.info/v1 and swapping in your unique API key.

For teams preferring raw HTTP requests, the standard POST method accepts JSON payloads containing your text, voice selection, and desired format. This flexibility allows seamless integration with any programming language or framework your startup already uses.

OpenAI SDK Compatibility

The API uses the same request shape as the OpenAI speech endpoint. By pointing your OpenAI SDK client to https://api.api36.info/v1 and providing your tts-... key, you can generate audio with minimal code changes.

  • Supports tts-1 and tts-1-hd models
  • Compatible with official OpenAI SDKs

Standard HTTP Requests

You can also send raw JSON POST requests directly to https://api.api36.info/v1/audio/speech. This method requires no external SDK dependencies, making it ideal for lightweight microservices or simple backend scripts in your course platform.

API Rate Limits

Each API key is limited to 60 requests per minute once your account is active. New accounts start with a lower limit of 10 requests per minute until the first top-up is processed.

  • 60 requests/minute for standard accounts
  • 10 requests/minute for new accounts

Audio Formats and Quality

Sound wave illustration for: Audio Formats and Quality

Selecting the correct audio format balances quality, file size, and playback compatibility. Our API supports MP3, WAV, Opus, FLAC, and PCM, allowing you to optimize delivery for different learning environments.

MP3 vs WAV vs Opus

MP3 offers the best balance of small file size and wide compatibility, ideal for web-based course players. WAV provides uncompressed, studio-quality audio suitable for high-fidelity listening but results in larger downloads. Opus delivers excellent quality at very low bitrates, reducing bandwidth costs without noticeable loss in speech clarity.

  • MP3: Best for general web playback and smaller downloads.
  • WAV: Uncompressed, highest fidelity, larger file sizes.
  • Opus: Efficient compression, ideal for low-bandwidth environments.

File Size Considerations

Larger files may slow down course loading, especially on mobile devices. Choose MP3 or Opus for standard lessons to ensure quick start times. Reserve WAV or FLAC for premium, high-fidelity modules where audio quality is the primary selling point.

Streaming vs Download

Our API returns complete audio files rather than streaming audio chunks. This means the entire file is delivered upon request completion, ensuring integrity but requiring the full file to be available before playback can begin. This approach simplifies integration for standard LMS players that expect a finished media file.

Handling Long Text Content

Sound wave illustration for: Handling Long Text Content

Long lesson scripts often exceed the API’s single-request limit. Understanding how the system handles larger volumes of text ensures your automated course production pipeline runs smoothly without interruption or data loss.

By breaking down lengthy content into manageable chunks, you can maintain audio continuity and keep processing times predictable for your students' on-demand playback.

Character Limits

Each API request accepts a maximum of 4,096 characters. This limit includes spaces and punctuation, so a dense lecture script may require multiple calls to fully convert a single module.

If your input exceeds this threshold, the request will fail unless you pre-split the text into smaller segments.

  • Maximum 4,096 characters per request
  • Spaces and punctuation count toward the limit

Splitting and Joining

Your application should divide the source text into parts of up to 4,096 characters each. After receiving the individual audio files from the API, your backend must join them into a single continuous audio file for the final course asset.

This approach ensures that no content is truncated while staying within the technical constraints of the endpoint.

  • Split text into ≤4,096 character chunks
  • Concatenate returned audio files sequentially

Processing Time

Generation speed is roughly real-time, averaging about 25 characters of text per second of audio. For short phrases, synthesis may take a few seconds, but longer texts process proportionally.

Requests are processed sequentially, so under heavy load, a request may wait in a queue before synthesis begins.

  • Approx. 25 characters per second of audio
  • Sequential processing with potential queue delays

Data Privacy and Usage

Sound wave illustration for: Data Privacy and Usage

EduEcho prioritizes your intellectual property by ensuring that input text is used exclusively for generating the immediate audio output. We do not use your lesson scripts to train our underlying models, preserving the uniqueness of your course content.

Text Not Used for Training

Your source text remains private. Unlike some services that incorporate user data into model improvement, we isolate your inputs strictly for synthesis purposes.

No Audio Storage

The generated audio files are returned directly in the API response and are not stored on our servers. Once you receive the file, we retain no copy of your final product.

Billing Metadata

We retain minimal request metadata—specifically timestamp, character count, and status—solely for accurate billing and usage tracking. This data does not include the text content or the audio itself.

Customer Support Options

Sound wave illustration for: Customer Support Options

You can reach the team via email for any account or technical questions. There is no phone number or live chat option, so responses are asynchronous. This approach keeps operations lean and costs low for startups.

Support is focused on resolving issues efficiently without the overhead of real-time channels. Check the documentation first for common integration patterns and voice selection tips.

Email Support

Send your queries to the support email address listed on the contact page. Since there is no live chat, allow 24-48 hours for a response during business days. Email is the sole channel for account and billing inquiries.

None

  • Sole support channel
  • No phone or chat
  • Asynchronous responses

Documentation Quality

The docs provide clear curl examples and SDK instructions for OpenAI compatibility. They cover all endpoints, including voice lists and language codes, to help you integrate quickly without guessing parameters.

None

  • Curl examples included
  • SDK instructions
  • Endpoint details

Community Resources

While there is no dedicated forum or community hub, you can find usage tips in the public API documentation. Developers often share integration snippets online, but official resources are limited to the docs and email support.

None

  • No dedicated forum
  • Public docs only
  • External snippets

Scalability for Growing Courses

Sound wave illustration for: Scalability for Growing Courses

As your course library grows, the API handles volume by processing requests sequentially. This means synthesis waits in line rather than failing, ensuring no data is lost during spikes.

You can scale usage by monitoring your 60 requests-per-minute limit. If you exceed it, requests queue automatically, preserving your billing integrity without extra complexity.

Handling High Volume

The system queues requests when the 60-per-minute limit is reached. Each request is processed in order, ensuring steady output even during peak lesson uploads.

Long texts are split into 4,096-character chunks and joined by your application. This allows you to generate full course modules without manual intervention or text truncation.

  • Sequential processing ensures no data loss during high traffic.
  • Automatic splitting handles texts larger than 4,096 characters.

Queue Management

Requests are processed one after another. Under heavy load, a single request may wait in the queue before synthesis begins.

Failed requests are free, so queue retries do not incur unnecessary costs. This makes the system resilient during high-volume testing phases.

Consistent Performance

Generation speed is roughly real-time for long texts (25 characters per second). Short phrases may take a few seconds due to queueing.

Performance remains consistent because the backend processes requests linearly. There are no variable latency spikes, only predictable wait times during high demand.

Common Mistakes to Avoid

Sound wave illustration for: Common Mistakes to Avoid

Avoiding common pitfalls ensures your online course audio generation remains smooth and cost-effective. Most errors stem from ignoring API constraints or making incorrect assumptions about language handling.

By understanding these limits upfront, you can prevent wasted credits and delivery delays during peak production times.

Ignoring Rate Limits

The API allows 60 requests per minute per key. Exceeding this limit causes delays as requests queue.

Startups often underestimate volume when generating full lesson series. Monitor usage to stay within limits.

  • Limit: 60 requests/minute
  • Pre-first top-up: 10 requests/minute

Not Setting Language Code

If the language field is omitted, the system guesses the language. Latin script defaults to English.

Non-English text in Latin script requires explicit language setting to ensure correct pronunciation.

  • Default: English for Latin script
  • Set 'language' for non-English

Overlooking Character Count

Each request supports up to 4096 characters. Spaces and punctuation count toward this limit.

Long scripts must be split into parts and joined to avoid truncation errors.

  • Max: 4096 characters/request
  • Includes spaces/punctuation

Final Recommendations for Startups

Sound wave illustration for: Final Recommendations for Startups

EduEcho’s TTS API is designed to streamline content production for online courses and LMS platforms. By converting lesson scripts into professional audio files programmatically, it simplifies the workflow for startups building scalable educational content.

Start with Free Trial

Begin with the $0.50 free trial (about 50,000 characters) to evaluate voice quality. This credit lasts for seven days, allowing you to test the API without immediate payment commitment. Note that trial credit is only available when signing in with Google.

  • Use the trial to test all 8 voice presets
  • Verify integration with your LMS platform
  • Check audio quality across different devices

Test with Real Content

Test the API with your actual lesson scripts to ensure tone and pacing match your brand. Use the 4096-character limit per request and split longer texts if needed. Experiment with different voices like Aria for narrators or Felix for energetic promos.

  • Test various voice presets with your content
  • Verify language detection accuracy
  • Check audio format compatibility

Monitor Usage Closely

Keep an eye on your character count and rate limits. The API allows 60 requests per minute per key. Failed requests are free, but successful ones are charged at $0.01 per 1,000 characters. Monitor your balance and usage through the dashboard.

  • Track character usage against your balance
  • Monitor rate limits to avoid delays
  • Set up alerts for low balance

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. These voices are best for core curriculum content where clarity and steady pacing are essential for retention.

Avatar of the voice aria
ariafemale
Avatar of the voice nova
novafemale
Avatar of the voice luna
lunafemale
Avatar of the voice sage
sagefemale
Avatar of the voice felix
felixmale
Avatar of the voice milo
milomale
Avatar of the voice atlas
atlasmale
Avatar of the voice orion
orionmale

23 languages

Any voice speaks any of these languages.

Sources and further reading

  1. Wikipedia
  2. Python documentation
  3. MDN: Web Speech API
  4. IETF RFC 7845: Ogg Opus
  5. Tether (USDT)
  6. LibriVox public-domain audiobooks
  7. openai-python on GitHub
  8. curl documentation
  9. W3C WAI: making audio and video accessible
  10. FFmpeg documentation
  11. ISO 639 language codes
  12. OpenAI API reference: create speech

More on this site