TTS API for Online Courses: Voices, Price & Limits

By CourseVoice editorial team · Updated 2026-11-19

This guide breaks down the CourseVoice API for engineering teams building LMS and e-learning platforms. We cover voice options, language support, pricing, and technical limits with concrete data from our specifications. You will see exactly how to integrate synthetic voice audio into your course engine using a standard HTTP endpoint compatible with OpenAI SDKs.

We address integration ease, performance metrics, and scalability to help you decide if this API fits your startup’s infrastructure. All information is specific to shweohh.info and based on current product facts.

TTS API for Online Courses: Voices, Price & Limits

Frequently asked questions

Which TTS API is best for LMS?

Q: Which TTS API is suitable for LMS? A: CourseVoice API is suitable for LMS platforms. It offers a simple POST endpoint compatible with OpenAI SDKs, supporting up to 4096 characters per request and 23 languages. The API returns audio files directly, allowing seamless embedding in course engines without complex infrastructure. It lacks streaming, SSML, and custom voice cloning, but provides reliable synthetic audio generation for educational content.

How do voices differ?

Eight presets cover distinct tones: female voices like Aria (warm narrator) and Nova (bright), and male voices like Felix (energetic) and Atlas (neutral). All voices support any of the 23 listed languages. Generated audio includes an inaudible watermark to identify synthetic speech.

Which language has the best quality?

Q: Which language has the highest quality? A: Quality is consistent across all 23 supported languages. Each voice speaks any supported language with equal fidelity, so no single language is superior. Voices are based on licensed recordings from LibriTTS-R and LibriVox, ensuring uniform synthetic speech quality regardless of the target language.

Is this API cheaper than Google?

Pricing is $10 per 1 million characters, or $0.01 per 1,000 characters. Failed requests are free. Costs are predictable based on character count. Unlike Google, there are no monthly plans or subscriptions, only pay-as-you-go credit via USDT top-ups starting at $5.

Does it support SSML?

No, the API does not support SSML. It accepts plain text input up to 4096 characters. You can adjust speed between 0.25 and 4.0 and select from eight voice presets. Complex audio formatting must be handled by your application logic.

Can I clone voices?

No, custom voice cloning is not offered. You must choose from the eight available presets: Aria, Nova, Luna, Sage, Felix, Milo, Atlas, or Orion. These are based on licensed recordings and cannot be customized or cloned from user-uploaded audio files.

How fast is the generation?

Generation is roughly real-time, processing about 25 characters per second. Short phrases may take a few seconds. Requests are processed sequentially, so under load, a request may wait in a queue. The API returns the complete audio file upon completion.

Is there a free tier?

Google sign-up accounts receive $0.50 trial credit, valid for 7 days, covering approximately 50,000 characters. Email/password accounts start with zero balance. There is no permanent free tier; you must top up with USTRC-20 tokens to continue usage after credits expire.

What formats are supported?

The API supports MP3, WAV, OPUS, FLAC, and PCM output formats. You specify the format in the request body. This allows flexibility for different LMS players and storage requirements. MP3 is suitable for most web-based course players due to its small file size.

Do I need a developer?

Yes, a developer is needed to integrate the API. You must send a POST request with JSON body containing model, input, voice, and format. The endpoint is compatible with OpenAI SDKs by changing the base URL. Basic HTTP knowledge is sufficient for integration.

Is the API reliable?

The API processes requests sequentially and may queue them under load. Rate limits are 60 requests per minute per key. Failed requests are free. There is no SLA or uptime guarantee. Support is available via email only, with no phone or live chat.

How do I compare prices?

Compare based on the $10 per 1 million characters rate. Count all characters, including spaces and punctuation. Other services may charge differently per second or word. CourseVoice API offers predictable costs with no hidden fees, and failed requests incur zero charge.

What is the character limit?

Each request supports up to 4096 characters. For longer texts, split the input into parts, generate audio for each, and join the files. This ensures consistent quality and manageable processing times. The API does not handle long-text splitting automatically.

Can I use it for video?

Yes, you can generate audio files for video narration. Supported formats include MP3, WAV, and OPUS. The audio can be synced with video editors. The API returns standard audio files that are compatible with most video production workflows and LMS video players.

Voice Options Comparison

Sound wave illustration for: Voice Options Comparison

The API provides eight distinct voice presets, each designed for specific instructional or narrative contexts in e-learning. These voices cover a range of tones, from warm narrators to energetic promo styles, ensuring your course audio matches the content’s intent. Each voice is based on licensed recordings, providing a consistent and professional baseline for synthetic speech.

All eight voices support all 23 available languages, allowing you to mix and match voices across different modules without language constraints. This flexibility simplifies localization for global LMS platforms.

Number of voices

The platform offers exactly eight voice presets. This limited set reduces decision fatigue while ensuring high quality across all options. Each voice is optimized for clarity and natural pacing, suitable for long-form course narration or short quiz prompts.

  • 8 total voice presets
  • Consistent quality across all voices
  • No custom voice cloning available

Voice types

Voices are categorized by tone and use case. Female voices include Aria (warm narrator), Nova (upbeat), Luna (storytelling), and Sage (instructional). Male voices include Felix (energetic promo), Milo (conversational), Atlas (neutral news), and Orion (deep documentary). This variety allows you to differentiate between lesson types, such as using Sage for technical tutorials and Luna for case studies.

Each voice is based on licensed recordings: LibriTTS-R (CC BY 4.0), from LibriVox public-domain audiobook readings. You can select any voice for any language, enabling dynamic course structures without re-recording.

Voice quality

Audio is generated as synthetic speech with an inaudible watermark marking it as AI-generated. The voices are clear and natural, optimized for listening rather than human-like emotion. They lack SSML support.

Quality is consistent across all requests. The engine does not offer emotion tags or pronunciation lexicons, so voice selection is the primary lever for tone. Generated audio is returned directly in the response and not stored, ensuring no residual data issues.

Language Support

Sound wave illustration for: Language Support

CourseVoice API supports 23 languages, ensuring your e-learning content reaches a global audience without language barriers. Every voice can speak any supported language, allowing consistent audio style across diverse courses. If the language parameter is omitted, the system guesses the language from the script, defaulting to English for Latin text.

This flexibility simplifies integration for startups building multilingual LMS platforms. You can maintain a single voice identity across different course modules while catering to regional learners.

Total languages

The API supports 23 languages, covering major global markets and many regional dialects.

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

Popular languages

English, Spanish, German, French, and Japanese are fully supported with high-quality voice presets. These languages cover the majority of e-learning demand in Western and Asian markets.

You can use any of the 8 voice presets for these languages, ensuring consistent brand audio even when switching between English and German courses.

  • English, Spanish, French, German, Japanese, Chinese, Portuguese, Russian

Niche languages

Support includes less common languages like Swahili, Malay, and Greek, enabling niche course creators to serve specific regional audiences. Non-English Latin text should explicitly set the language parameter to avoid misidentification.

Languages like Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, and Bengali are not supported. Plan your content strategy accordingly if targeting these regions.

  • Swahili, Malay, Greek, Turkish, Hebrew, Hindi, Finnish, Danish

Pricing Structure

Sound wave illustration for: Pricing Structure

Our pricing is straightforward: $10 per 1 million characters, equating to $0.01 per 1,000 characters. Spaces and punctuation count toward the total. Failed requests are free, with charges returned in full if synthesis fails.

There is no minimum spend and no monthly plans. You only pay for what you generate. This model suits startups scaling their course libraries without committing to fixed infrastructure costs.

Cost per unit

Each character costs $0.00001.

The rate applies to all models, including tts-1 and tts-1-hd, which share the same engine and price point.

  • $10 per 1 million characters
  • $0.01 per 1,000 characters
  • Failed requests are free

Minimum spend

There is no minimum spend requirement. You can top up any amount, starting from $5.

Balance does not expire, allowing you to store credits for future course updates or new modules.

  • No minimum spend
  • No monthly plans
  • No subscriptions

Trial options

Sign in with Google to receive $0.50 in trial credit, valid for 7 days. This covers about 50,000 characters.

Email-based accounts start without trial credit. Top up via USDT on the TRON network to begin using the API immediately.

  • $0.50 trial credit
  • Valid for 7 days
  • Google sign-in required

Integration Ease

Sound wave illustration for: Integration Ease

Integrating CourseVoice API into your LMS requires minimal overhead. Since the endpoint mirrors the OpenAI speech format, you can reuse existing HTTP clients or the official OpenAI SDK by simply pointing it to https://api.shweohh.info/v1. This eliminates the need for custom parsers or complex serialization logic in your course engine.

The API complexity is low because it uses standard JSON bodies and returns audio files directly. You do not need to manage WebSocket connections or handle partial streaming responses, keeping your backend logic straightforward and predictable.

SDK support

There is no proprietary SDK. Use the official OpenAI SDKs or any standard HTTP library. Change the base_url to https://api.shweohh.info/v1 and use your API key. This works with Python, Node.js, and other languages immediately.

This approach ensures compatibility with the tools your engineering team already uses for LMS development.

  • Official OpenAI SDKs compatible via base_url change
  • Works with any standard HTTP client
  • No custom client library required

API complexity

The endpoint POST /v1/audio/speech accepts simple JSON: model, input text, voice, and optional parameters like speed or language. Responses are standard audio files (mp3, wav, etc.).

Error handling is straightforward: failed requests return standard HTTP error codes without charging your balance, simplifying retry logic in your course player.

Documentation

Complete endpoint details, voice lists, and usage examples are available at /docs/. The documentation includes ready-to-use curl commands.

This reduces onboarding time for developers who need to connect synthetic voice audio to course video players or text-to-speech modules quickly.

Performance Metrics

Sound wave illustration for: Performance Metrics

Synthesis performance depends on text length and server load. Long-form content generates at roughly real-time speed, processing approximately 25 characters per second. Short phrases require a few seconds to complete.

Generation speed

Processing occurs sequentially, meaning requests are handled one after another. Under heavy load, your request may wait in a queue before synthesis begins. This ensures consistent output quality but affects total turnaround time for bulk uploads.

  • ~25 characters generated per second of audio
  • Sequential processing under load
  • Queue delays possible during peak usage

Latency

Latency is driven by the queue position and engine processing time. There is no streaming or partial audio delivery. You receive the complete audio file only after synthesis finishes, which takes a few seconds for short texts and longer for full chapters.

Reliability

Requests are processed in order. If synthesis fails, the charge is returned in full, ensuring no cost for wasted effort. The system does not guarantee specific uptime or SLA, so plan for occasional delays during high-volume periods.

  • Full refund on failed synthesis
  • No SLA or uptime guarantee
  • Sequential request handling

Use Case Fit

Sound wave illustration for: Use Case Fit

Synthetic voices eliminate the need for expensive studio bookings when scaling course content. You can generate consistent narration for hundreds of lessons using a single API call, ensuring uniform audio quality across your entire LMS.

This approach allows rapid iteration on script changes without re-recording entire modules, significantly reducing production overhead for growing ed-tech platforms.

Course narration

Use the API to convert script text into MP3 or WAV files for your learning management system. With voices like Aria or Felix, you can maintain a consistent brand voice across all modules. The 4096-character limit per request suits typical lesson segments well.

You can switch voices between modules to keep learners engaged, using Sage for instructional content or Orion for documentary-style deep dives.

  • Consistent audio quality across all courses
  • Easy script updates without re-recording
  • Support for multiple tones and styles

Accessibility

Generate audio versions of text-based lessons to support diverse learning needs. The API supports 23 languages, enabling global accessibility.

The inaudible watermark confirms synthetic origin, maintaining transparency while delivering high-quality listening experiences.

  • Support for 23 languages
  • WCAG compliance enhancement
  • Transparent synthetic audio

Video content

Sync generated audio with video lessons for a complete multimedia experience. Use the speed parameter to adjust pacing for different audience preferences. The API returns audio files ready for integration with your video hosting platform.

This method reduces production time significantly compared to traditional voiceover processes, allowing faster content rollout.

  • Sync with video lessons
  • Adjustable speed for pacing
  • Fast content rollout

Technical Limits

Sound wave illustration for: Technical Limits

The API enforces a strict 4,096-character limit per request, including spaces and punctuation. For longer course modules, split the text into chunks and concatenate the resulting audio files. Short phrases typically take a few seconds to generate, while longer texts process at roughly real-time speed (25 characters per second).

Requests are processed sequentially, meaning high traffic may introduce queue latency. The system handles concurrency by queuing requests, ensuring no data loss but potentially increasing wait times during peak usage.

Character limit

Each API call accepts up to 4,096 characters of input text. This limit applies to the entire string sent in the input field. If your course content exceeds this, you must split the text into smaller segments and join the generated audio files on the client side or via a backend script.

Rate limits

Standard accounts are limited to 60 requests per minute per API key. New accounts start with a lower limit of 10 requests per minute until the first top-up is processed. Requests are processed one after another, so under load, a request may wait in a queue before synthesis begins.

  • 60 requests/minute (standard)
  • 10 requests/minute (new accounts)
  • Sequential processing may cause queuing delays

File formats

The API supports five output formats: mp3, wav, opus, flac, and pcm. Choose based on your LMS requirements: mp3 for web delivery, wav for high-quality local playback, or pcm for raw data processing. The generated audio includes an inaudible watermark to identify synthetic speech.

  • mp3
  • wav
  • opus
  • flac
  • pcm
Technical Limits

Payment Methods

Sound wave illustration for: Payment Methods

CourseVoice API uses a single payment method for simplicity. All top-ups are processed exclusively via USDT on the TRON network (TRC-20). There are no credit cards, no PayPal, and no subscription models to manage. This approach keeps overhead low and aligns with the startup-friendly, engineering-first ethos of our platform.

You can add funds to your balance at any time. The minimum top-up amount is $5, and credits are typically available within minutes of your transfer confirmation. Your balance never expires, allowing you to pay as you grow without worrying about monthly cycles or unused credits.

Crypto only

We accept USDT on the TRON network (TRC-20) for all top-ups. This single-crypto model reduces transaction fees and simplifies your accounting. Credits are credited automatically within minutes after the blockchain confirms your transfer. There are no other payment gateways or middlemen involved in the funding process.

  • USDT on TRON (TRC-20) only
  • Minimum top-up: $5
  • Automatic credit within minutes

Credit cards

We do not process credit cards directly. By avoiding traditional payment processors, we eliminate recurring subscription fees and complex billing cycles. You only pay for what you use, with no monthly minimums or seats to manage.

Invoices

We do not issue invoices or handle VAT. This eliminates administrative overhead for both you and us, keeping the focus on delivering audio.

Data Privacy

Sound wave illustration for: Data Privacy

We prioritize data integrity for educational content. Your input text is processed solely to generate the target audio file and is not used for model training. This ensures your course materials remain unique to your platform without contributing to our training dataset.

Generated audio is returned directly in the API response and is not stored on our servers. We retain only request metadata—such as timestamp, character count, and status—strictly for billing purposes. This minimal retention policy keeps your usage data lean and auditable.

Training data

Input text is used exclusively for synthesis and is not retained for training. This guarantees that your specific course content does not influence future model iterations.

Unlike many consumer TTS providers, we do not repurpose your uploaded scripts to improve our base models.

Storage policy

Audio files are delivered in the response and not persisted on our infrastructure. Only request metadata (time, characters, status) is kept for billing accuracy.

This ephemeral storage model reduces data liability and simplifies compliance for educational startups handling student-generated content.

Watermarks

Every generated audio file contains an inaudible watermark identifying it as synthetic speech. This helps distinguish AI-narrated lessons from human-recorded ones without affecting audio quality.

The watermark is embedded at the audio level, ensuring it persists through standard format conversions like MP3 or WAV.

Support Level

Sound wave illustration for: Support Level

Support for CourseVoice API is handled exclusively via email. There is no phone line, live chat widget, or ticketing portal. This keeps overhead low, which helps maintain our simple pricing model without hidden service fees.

All technical questions, billing issues, and integration help requests go to our support address. We do not offer priority lanes or dedicated account managers for startups.

Email only

We rely on asynchronous email communication for all support needs. This approach allows our engineers to review code snippets and error logs thoroughly before responding, rather than rushing through live chats.

Please include your API key and the request ID from the response header for faster resolution.

  • No phone support
  • No live chat
  • No premium support tiers

Documentation

Our documentation is comprehensive and directly linked from the /docs/ page. It includes code examples for Python, Node.js, and cURL, along with a full reference of the /v1/audio/speech endpoint.

The docs cover authentication, rate limits, and error codes. They are written for developers integrating with LMS platforms, so you can find the exact payload structure you need without guessing.

Response time

While we do not guarantee a specific response window, we aim to reply within one business day.

For urgent billing issues regarding your TRON network top-ups, please mark your email as urgent so our team can verify the transaction hash directly.

Scalability

Sound wave illustration for: Scalability

The API handles standard batch loads well, but request concurrency is constrained by a 60 requests-per-minute limit per key. This ceiling suits most course production workflows without requiring complex sharding.

Request volume

Each key processes up to 60 requests per minute. For large course libraries, split generation jobs across multiple keys or batch smaller segments to stay within this limit without throttling.

  • 60 requests/minute per key
  • 10 requests/minute before first top-up

Queue handling

Requests are processed sequentially. Under heavy load, synthesis may wait in a queue, increasing latency. Short phrases take a few seconds; long texts generate at roughly real-time speed.

Growth potential

Scale by issuing additional keys for parallel processing. Since balance never expires and top-ups are instant via USTRC20, you can ramp volume immediately without waiting for invoice approvals or credit checks.

Final Recommendation

Sound wave illustration for: Final Recommendation

This API fits startups building LMS platforms who need reliable synthetic voice generation without complex infrastructure. It delivers clear, natural-sounding audio via a simple POST endpoint, ideal for course narration and accessibility features.

The combination of low cost, straightforward integration, and transparent limits makes it a practical choice for early-stage edtech products scaling their content library.

Best for startups

Startups benefit from the zero-infrastructure model and predictable pricing. The API handles voice synthesis without requiring you to manage servers or voice licenses.

It integrates directly with your course engine, allowing rapid iteration on audio content without heavy engineering overhead.

  • No infrastructure management
  • Rapid content iteration
  • Simple POST endpoint

Cost effective

At $0.01 per 1,000 characters, costs scale linearly with usage. No monthly fees or minimum spend required.

Failed requests are free, and the free trial provides initial credit for testing.

  • $0.01 per 1,000 characters
  • No monthly fees
  • Pay-as-you-go model

Easy integration

Uses the standard OpenAI SDK shape, so existing integrations work with a simple base_url change.

Comprehensive documentation and ready-to-use cURL examples accelerate development time.

  • OpenAI SDK compatible
  • Standard JSON body
  • Ready cURL examples

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. You only pay for what you use, with no monthly minimums or seats to manage.

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. Node.js documentation
  3. ISO 639 language codes
  4. LibriVox public-domain audiobooks
  5. LibriTTS-R paper (arXiv 2305.18802)
  6. curl documentation
  7. MDN: Web Speech API
  8. OpenAI API reference: create speech
  9. TRON developer hub: TRC-20
  10. IETF RFC 7845: Ogg Opus
  11. Creative Commons Attribution 4.0
  12. Tether (USDT)

More on this site