Frequently asked questions
What is an LMS?
An LMS (Learning Management System) is software for delivering educational content. Our API generates audio files that you can upload directly to your LMS. This allows you to convert course scripts into listenable audio without complex infrastructure, making your content accessible to learners who prefer listening.
Can I use this API with Moodle?
Yes. Moodle supports standard audio file formats. Generate your audio via our API, then upload the resulting file (mp3, wav, opus, flac, or pcm) as an audio resource or in a SCORM package. Since Moodle accepts standard web audio formats, our output integrates seamlessly into your course structure.
How do I upload audio to Canvas?
Canvas accepts common audio formats. After generating audio via our API, download the file and upload it as a media file or within an assignment. Canvas supports mp3, wav, opus, flac, and pcm, so any format from our API works directly without conversion.
Does the API support SCORM?
The API returns raw audio files, not SCORM packages. However, you can generate the audio via our API and then bundle it into a SCORM package using your LMS authoring tool or a third-party converter. The audio itself is standard and SCORM-compatible.
Can I automate audio generation?
Yes. Use the POST endpoint to send text and receive audio programmatically. You can script this using the OpenAI SDK or plain HTTP. Automate the process by sending course text via API, receiving the audio file, and uploading it to your LMS via its API.
What is the best format for LMS?
MP3 is a widely supported format for LMS platforms. Our API also supports WAV, OPUS, FLAC, and PCM. MP3 offers the best balance of quality and file size for web delivery, ensuring compatibility across most learning management systems and devices.
How do I handle long lessons?
Split text into parts of up to 4096 characters. Generate audio for each part via separate API requests, then join the audio files. This ensures reliable synthesis and avoids exceeding the character limit per request, allowing you to create complete lesson audio files.
Can I change voices later?
Yes. Generate new audio with a different voice preset by changing the voice parameter in your API request. Since audio files are not stored by the API, you simply regenerate the content with the new voice and re-upload it to your LMS.
Is the audio MP3 compatible?
Yes. MP3 is one of the five supported response formats. It is widely compatible with major LMS platforms, including Moodle, Canvas, and Blackboard. Use MP3 for broadest compatibility, or choose WAV, OPUS, FLAC, or PCM if your LMS supports those.
How do I track costs per course?
Track characters used per course. Our pricing is $10 per 1 million characters. Monitor your usage in the dashboard to see characters consumed. Multiply the character count by $0.00001 to calculate the exact cost for each course’s audio generation.
What if the API is down?
We do not offer an SLA or uptime guarantee. If the API is unavailable, requests may fail. Failed requests are free, so you are not charged for failed attempts. Retry your requests when the service is back online.
Do I need a developer?
No. You can use the API with no-code tools or simple scripts. The endpoint matches the OpenAI SDK, so you can use existing integrations. Many LMS platforms allow file uploads, so you can generate audio externally and upload it manually.
Can I use this for podcasts?
Yes. Generate audio files via the API and download them. You can then upload these files to your podcast hosting provider. The API supports standard formats like MP3 and WAV, which are compatible with all major podcast platforms.
How do I test the API?
Use the free trial credit of $0.50. Sign in with Google to receive this credit. Send a test request via the API endpoint with sample text. Check the response for the audio file. If it fails, no charge is applied.
Prepare your course script
Start by exporting your course script directly from your LMS. Most platforms allow you to download lesson text as plain text files or CSVs. This ensures you have a clean, editable source file ready for processing.
Before sending data to the API, verify the length of your script. The API accepts up to 4,096 characters per request. Knowing your character count helps you plan how many API calls you will need for each lesson module.
Export text from LMS
Access your LMS course editor and locate the text content for each lesson. Use the platform’s built-in export feature to save scripts as .txt or .csv files. This creates a reliable local copy that you can edit before generating audio.
Ensure the exported text is clean and free of HTML tags or formatting codes that might confuse the API. Simple plain text works best for the input field in your API request.
- Use your LMS’s native export tool
- Save as .txt or .csv format
- Remove HTML tags if present
Check character count
Count every character in your script, including spaces and punctuation. The API charges $0.01 per 1,000 characters, so accurate counting helps you predict costs. Remember that one request can handle up to 4,096 characters.
If your script exceeds 4,096 characters, you will need to split it. Use a simple text editor or script to divide the text into smaller chunks that fit within this limit.
- Count all characters including spaces
- Limit is 4,096 characters per request
- Spaces and punctuation count
Split long texts
Divide long scripts into segments of no more than 4,096 characters. Ensure splits occur at natural breaking points, such as the end of a sentence or paragraph, to avoid awkward audio pauses. Generate each segment as a separate audio file.
After generating audio for each part, you can join the files using any standard audio editor or script. This allows you to create a single, continuous audio file for your LMS lesson, maintaining a smooth student experience.
- Split at sentence boundaries
- Keep segments under 4,096 chars
- Join files for continuous playback
Generate audio files
To generate audio, send your course script to the POST https://api.play66f.com/v1/audio/speech endpoint. The API accepts JSON bodies with your text, chosen voice, and preferred format. Each request handles up to 4096 characters, so split longer lesson scripts into smaller parts before sending.
Use the API endpoint
Send a POST request to https://api.play66f.com/v1/audio/speech with a JSON body containing the input text, model, voice, and response_format. Include your API key in the Authorization: Bearer tts-... header. The endpoint processes requests sequentially, so under load, synthesis may queue briefly.
- Endpoint:
POST /v1/audio/speech - Auth:
Bearer tts-... - Max input: 4096 characters
Select voice and format
Choose from eight preset voices like aria for narrators or atlas for news-style lessons. Select a response format such as mp3, wav, opus, flac, or pcm. Adjust the speed parameter between 0.25 and 4.0 to match your course’s pacing requirements.
- Voices: aria, nova, luna, sage, felix, milo, atlas, orion
- Formats: mp3, wav, opus, flac, pcm
- Speed: 0.25–4.0
Download MP3 or WAV
The API returns the audio file directly in the response. MP3 is ideal for LMS compatibility and smaller file sizes, while WAV offers lossless quality. Failed requests incur no charge, ensuring you only pay for successfully generated lesson audio.
Upload to your LMS
Once your audio files are generated, the next step is integrating them into your Learning Management System. Most LMS platforms allow you to attach media files directly to lesson pages or embed them within text blocks. Simply upload the MP3 or WAV file to your course media library, then insert the audio player into the specific lesson where it belongs.
Ensure each audio file is clearly named to match its corresponding lesson module, making it easy to locate and update later. This manual step ensures precise control over which audio accompanies each learning objective.
Add audio to lessons
Navigate to your LMS lesson editor and use the built-in media uploader to add your synthesized audio file. Most platforms support standard formats like MP3 and WAV, which are compatible with nearly all student devices. Upload the file directly from your computer or import it from a connected cloud storage service.
Assign a clear filename that reflects the lesson content, such as "lesson-1-intro.mp3," to maintain organization. This makes future updates or replacements straightforward without breaking existing links.
- Upload MP3 or WAV files via the LMS media library
- Use descriptive filenames for easy identification
- Ensure compatibility with common student devices
Link files to modules
After uploading, link the audio file to the specific lesson module in your course structure. This ensures students access the correct audio when they engage with the corresponding text or video content. Some LMS platforms allow you to embed audio players directly into lesson pages, while others may require manual linking via URL.
Verify that the audio file is publicly accessible or appropriately permissioned for student viewing. If your LMS requires a specific embed code, copy and paste it into the lesson editor to ensure seamless playback for learners.
- Embed audio players directly into lesson pages
- Verify public access permissions for student playback
- Use embed codes if required by your LMS
Test playback
Before launching your course, test the audio playback on multiple devices and browsers to ensure compatibility. Play the audio file within the LMS to confirm it starts, pauses, and stops correctly. Check for any latency issues or formatting problems that might distract students.
Review the audio quality to ensure it matches the generated file’s specifications. If any issues arise, re-upload the corrected file or adjust the LMS settings to resolve playback errors. This step ensures a smooth learning experience for all users.
- Test on multiple devices and browsers
- Verify start, pause, and stop functionality
- Check for latency or formatting issues
Automate with scripts
Automate your TTS workflow using the OpenAI SDK by pointing it to our base URL. This allows you to generate audio files programmatically without writing custom HTTP client code. You can integrate this directly into your course creation pipeline to convert lesson scripts into audio on demand.
Batch processing lets you convert multiple lesson scripts efficiently. By splitting long texts into 4096-character segments, you ensure consistent audio quality while managing API limits. Use simple scripts to loop through your content library, generating audio files that are ready for immediate upload to your LMS.
Use OpenAI SDK
Point the OpenAI SDK to https://api.play66f.com/v1 and use your API key. This approach leverages familiar libraries for Python or Node.js, making integration straightforward for no-code developers who know basic scripting.
The SDK handles the JSON structure for POST /v1/audio/speech, ensuring correct formatting for model, input, and voice parameters. This reduces boilerplate code and minimizes integration errors.
- Install the standard
openaipackage. - Set
base_urlto our API endpoint. - Pass your
tts-...key in the header.
Batch processing
Process multiple lessons by iterating through your script list. For texts exceeding 4096 characters, split the input, generate each segment, and concatenate the resulting audio files. This maintains audio continuity for longer course modules.
Batch jobs can be scheduled to run overnight, preparing all audio assets for the next day’s lessons. This reduces manual effort and ensures consistent voice output across your entire course library.
- Loop through lesson scripts.
- Split text at 4096 characters.
- Concatenate generated audio segments.
Error handling
Implement retry logic for rate limit errors (60 requests per minute). If a request fails, the charge is returned, so you can safely retry without incurring extra costs. Handle network timeouts gracefully to avoid losing progress.
Log failed requests to identify issues like invalid voice IDs or unsupported languages. Proper error handling ensures that your automation script continues running even if individual lessons fail to generate.
- Retry on rate limits.
- Check for API errors.
- Log failed generations.
Manage voice consistency
Maintaining a consistent vocal identity across your course modules enhances student retention and professional polish. By assigning a specific voice to each course, you create an auditory brand that learners recognize instantly.
Use same voice per course
Assign a single voice, such as Aria or Felix, to all lessons within a specific course. This ensures tonal consistency, making the learning experience feel cohesive and professionally produced rather than disjointed.
Mix voices for characters
For courses with multiple perspectives, use different voices to distinguish speakers. For example, use Milo for conversational explanations and Atlas for news-style updates. This helps students mentally separate distinct topics or personas.
- Use distinct voices for narrator vs. interview subjects
- Differentiate between theory sections and practical examples
- Assign unique voices to different course modules
Store voice presets
Document your chosen voice for each course in your internal style guide. When generating new audio, simply reference the preset voice ID in your API request to ensure you never accidentally switch tones mid-course.
Keep a simple spreadsheet mapping course IDs to voice names like Luna or Orion. This reduces decision fatigue and speeds up bulk generation workflows for existing or new content.
- Record voice choices in a course style guide
- Map course IDs to specific voice presets
- Review voice assignments before bulk API requests
Handle language requirements
Language support is critical for diverse course audiences. Our API handles 23 languages, including English, Spanish, French, German, Chinese, Japanese, and others. Each voice can speak any supported language, ensuring consistent quality across your curriculum.
Set language code
Always specify the language parameter in your request using the ISO 639-1 code (e.g., "language": "fr" for French). This guarantees the engine applies the correct phonetics and intonation patterns, preventing awkward accents.
- Supported codes include
en,es,fr,de,zh,ja,ko,ru,ar,hi,it,pt,nl,sv,no,da,fi,pl,tr,el,he,ms,sw
Guess language from script
If you omit the language field, the system attempts to detect the language from the text script. Latin script text defaults to English if detection fails or is ambiguous. For non-English Latin-script text, always set the language explicitly to avoid mispronunciation.
Multi-language courses
You can mix languages within a single course by selecting the appropriate voice and language code for each segment. The API does not support custom voice cloning, so you must manually assign voices to match the target language of each lesson segment.
- Unsupported languages: Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, Bengali
Monitor usage and costs
Tracking your API usage ensures you stay within budget while generating course audio. The dashboard displays your remaining balance, character count, and recent activity, allowing you to anticipate costs before they accumulate.
Understanding how characters are counted helps avoid unexpected charges. Every character, including spaces and punctuation, contributes to the total. With a rate of $0.01 per 1,000 characters, you can easily estimate the cost of any script before sending it to the API.
Check character usage
Access your account dashboard to view real-time character consumption. Each API request includes an input field with a limit of 4,096 characters. Long course modules may require splitting text into multiple requests to stay within this limit, ensuring accurate billing per segment.
- View current balance and character count
- Monitor request history for accuracy
- Estimate costs before bulk generation
Track API calls
The API enforces a rate limit of 60 requests per minute per key. Exceeding this limit may result in temporary delays or errors. Monitor your request volume to ensure smooth playback generation, especially when uploading large course libraries.
- Rate limit: 60 requests per minute
- Queue processing for high volume
- Monitor request status codes
Top up balance
Your balance does not expire, and failed requests are refunded. To add funds, use USDT on the TRON network (TRC-20) with a minimum of $5. Transfers are credited automatically within minutes, ensuring uninterrupted service for your course production.
- Supports USDT (TRC-20) only
- Minimum top-up: $5
- Automatic credit within minutes

Troubleshoot common issues
If your audio files don't play in the LMS, verify the file format matches your platform's requirements. The API returns MP3, WAV, Opus, FLAC, or PCM files. Ensure your LMS is configured to accept these specific formats without additional conversion.
Check that the audio file is not corrupted during download. If the file size is unexpectedly small, the generation may have failed silently. Review your API response logs for any error codes.
Audio not playing
Ensure the LMS supports the chosen format like MP3 or WAV. If using Opus or FLAC, convert the file first.
Verify the file integrity by checking the file size against expected generation times.
- Check LMS format compatibility
- Verify file integrity
Incorrect voice
If the voice sounds wrong, confirm you selected the correct voice parameter in your request. The API supports 8 specific presets like Aria, Nova, or Felix.
Ensure the language parameter matches the script's language. If omitted, the API guesses the language, which may default to English for Latin script.
- Check voice parameter
- Set language code
Rate limit errors
If you receive a 429 error, you have exceeded 60 requests per minute. Wait before retrying.
For new accounts, the limit is 10 requests per minute until the first top-up.
- Wait for rate limit reset
- Check account tier limits
Optimize for student experience
Generating audio for short phrases takes only a few seconds, enhancing the learning flow by providing quick audio feedback without waiting for large batch jobs to complete. This allows creators to quickly iterate on script changes and hear the result instantly, ensuring the pacing matches the intended teaching style.
Audio quality directly impacts student retention. By selecting high-fidelity formats like WAV or FLAC, you ensure that instructional nuances are preserved. Clear, natural-sounding narration reduces cognitive load, helping learners focus on the course material rather than struggling to understand the voice.
Real-time generation
For short phrases, generation takes only a few seconds. Longer texts process at roughly real-time speed, allowing for near-instant playback in interactive modules.
Requests are processed sequentially, so under heavy load, a brief queue may occur. This ensures consistent audio quality without compromising the generation timeline.
- Natural pacing for most instructional content.
- Minimal delay for text under 4096 characters.
Audio quality
Choose WAV or FLAC for lossless quality, ensuring crisp narration for detailed explanations. MP3 is suitable for general lessons where file size matters more than studio-grade fidelity.
All audio files include an inaudible watermark, confirming the content is synthetic and protecting your original course structure without affecting the listening experience.
- WAV/FLAC for maximum clarity.
- MP3 for balanced quality and size.
- Inaudible watermark for authenticity.
File size management
Selecting the right format balances quality with bandwidth. FLAC and WAV files are larger but offer superior sound, while MP3 keeps file sizes manageable for faster loading on various devices.
Use PCM for raw data if your LMS handles compression, or stick to MP3 for broad compatibility. This ensures smooth playback even on slower connections.
- WAV/FLAC: High quality, larger files.
- MP3: Balanced size and quality.
- PCM: Raw data for custom compression.
Secure your content
Protect your course assets with built-in synthetic markers and strict data practices. Each generated audio file carries an inaudible watermark, clearly marking it as AI-generated speech. This helps maintain transparency with your students and provides a simple way to identify the source of the audio without altering the listening experience.
Watermarked audio
The watermark is embedded directly into the audio file during synthesis. It remains inaudible to listeners but can be detected to verify the content originated from our API. This adds a layer of authenticity without requiring external tools or metadata checks.
No data retention
Your input text is used solely to produce the audio and is not stored for training purposes. The generated audio is returned in the response and is not retained on our servers. Only request metadata, such as time and character count, is kept for billing accuracy.
Private API key
Each account uses a single active API key, kept private to your infrastructure. You can issue a new key at any time, which instantly invalidates the old one. This ensures that if a key is compromised, access can be revoked immediately without affecting other services.
- Revoke old keys instantly upon issuing a new one
- Use one key per account for simple tracking
- Keys start with
tts-for easy identification
Scale your course library
Scaling your course library requires processing many scripts efficiently. You can split long texts into 4096-character parts and join the resulting audio files to create complete lessons.
Batch processing allows you to generate multiple files in sequence. This ensures consistent quality across your entire library without manual intervention for each file.
Process multiple courses
Use the same API key for all courses. This simplifies billing and management. Each course script is processed independently, allowing you to generate audio for dozens of lessons in parallel.
- Use consistent voices per course
- Batch process multiple scripts
- Maintain uniform audio quality
Handle high volume
The API handles 60 requests per minute. For larger libraries, schedule generation during off-peak hours. This avoids delays and ensures all audio files are ready before student enrollment.
Queue management
Requests are processed sequentially under load. Use a simple queue system in your script. This prevents overwhelming the API and ensures no audio file is lost during high-volume generation.
Get support
Our support is straightforward and designed for independent creators. Since we do not offer phone lines, live chat, or SLAs, all assistance is handled via email. This keeps our operational costs low, allowing us to maintain our simple character-based pricing without hidden fees for premium support tiers.
Email support
For any technical questions or billing issues, send an email to our support address. We handle all inquiries directly, ensuring you get specific answers without navigating a complex help desk system. Please include your API key and a brief description of your issue for faster resolution.
Documentation
The most efficient way to solve integration issues is through our detailed API documentation. It covers all endpoints, request formats, and common error codes. You can also find ready-made code snippets for the OpenAI SDK, which simplifies the integration process for most LMS platforms.
- Full endpoint reference
- Code examples for Python and cURL
- Error code explanations
Community resources
While we do not manage a formal community forum, many no-code creators share their integration tips on social media and developer forums. Searching for 'SpeakFlow API' or our endpoint URL often reveals practical workflows from other course creators who have already solved similar challenges.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. This approach avoids the complexity of custom voice cloning while providing high-quality, consistent synthetic speech.
23 languages
Any voice speaks any of these languages.
