Frequently asked questions
What is text-to-speech technology?
Text-to-speech (TTS) converts written text into audible speech. Our API sends text and returns an audio file, allowing you to turn manuscripts into listenable audio without hiring voice actors.
How does a TTS API work?
Send a POST request to the speech endpoint with your text, voice, and format. The API processes the request and returns the generated audio file directly in the response.
Which voice is best for audiobooks?
Choose based on your story’s tone. Luna offers soft, storytelling narration, while Sage provides calm, instructional clarity. Aria is a warm narrator, and Orion suits deep, documentary-style content.
Can I use this for commercial projects?
Yes. The generated audio is returned in the response and not stored. Input text is used only for production. You can use the output for your commercial audiobook projects.
How long does it take to generate an audiobook?
Generation is roughly real-time, processing about 25 characters per second. Short phrases take a few seconds. Long texts are split and joined, so total time depends on length and queue load.
Do I need to hire a voice actor?
No. Our API generates natural-sounding narration in 23 languages using licensed voices. You can produce audiobooks directly from your manuscript without hiring human voice actors.
What is the difference between tts-1 and tts-1-hd?
Both models use the same engine and price. The only difference is the model identifier in your request. Use either model, as they use the same engine and price.
How do I split long texts?
Split your text into parts of up to 4096 characters each. Send each part as a separate request and join the resulting audio files to create your full audiobook.
Can I change the speed of the audio?
Yes. You can adjust the speed parameter between 0.25 and 4.0. This allows you to make narration faster or slower to fit your preferred pacing.
What languages are supported?
We support 23 languages including English, German, French, Spanish, Japanese, and Arabic. Any voice can speak any of these languages. Set the language code for non-English Latin-script text.
Is the audio royalty-free?
The voices are based on licensed recordings from LibriVox public-domain audiobook readings. The generated audio files are returned to you and not stored, allowing you to use them in your projects.
Do I own the generated audio?
The audio is returned in the response and not stored by us. Input text is used only for production. You retain control over the generated audio files for your projects.
How do I handle accents?
Select a voice that matches your desired accent. Each voice has a distinct tone. For non-English text, set the language parameter to ensure correct pronunciation for that specific language.
Can I combine multiple voices?
Yes. Generate audio for different sections using different voices. Split your text, request audio for each part with the desired voice, and then join the audio files together.
What is the character limit?
Each request accepts up to 4096 characters. For longer manuscripts, split the text into chunks of this size, generate each chunk, and join the resulting audio files.
How do I pay for the service?
Pay using USDT on the TRON network (TRC-20). Minimum top-up is $5. Credits are added automatically within minutes. There are no subscriptions or monthly plans.
Is there a free trial?
Yes. Sign in with Google to receive $0.50 of credit, valid for 7 days. Email-based accounts start without trial credit. This covers about 50,000 characters.
Can I use my own voice?
No. We do not offer custom voice cloning or uploading your own voice. You can choose from eight preset voices based on licensed recordings.
What happens if the service goes down?
We do not provide an SLA or uptime guarantee. If synthesis fails, the charge is returned in full. Requests are processed sequentially, so load may cause delays.
Do I need coding skills?
No. You can use the official OpenAI SDK by changing the base URL, or use plain HTTP requests. We also provide ready curl examples on the dashboard.
How do I get started?
Go to /get-api-key/ and sign up with Google or email. You will receive an API key. Use it with the OpenAI SDK or HTTP requests to generate audio.
What is a synthetic watermark?
Every generated audio file carries an inaudible watermark. This marks the file as synthetic speech, distinguishing it from human-recorded audio without affecting the listening experience.
Are the voices realistic?
The voices are based on licensed recordings from LibriVox public-domain audiobook readings. They offer natural-sounding narration suitable for self-publishing and audiobook creation.
Can I edit the audio after generation?
Yes. The API returns standard audio formats like mp3, wav, opus, flac, or pcm. You can edit these files using any standard audio editing software.
Is the API reliable?
We do not offer an SLA or uptime guarantee. Requests are processed sequentially. Under load, requests may wait in a queue. Failed requests are charged zero.
How do I contact support?
Support is available by email only. There is no phone number, live chat, or on-site contact form. Send your inquiries directly to our support email address.
What is USDT?
USDT is a stablecoin. We accept USDT on the TRON network (TRC-20) for top-ups. The minimum amount is $5, and credits are added automatically within minutes.
Can I get a refund?
Failed requests are free, and the charge is returned in full. For successful requests, refunds are not explicitly guaranteed beyond the failed request policy. Check the pricing page for details.
How do I rotate my API key?
You can issue a new API key at any time from your account dashboard. The old key stops working immediately when the new one is created.
What is the base URL?
The API base URL is https://api.44bd.info/v1. Use this URL with the OpenAI SDK or in your HTTP requests to access the speech and voices endpoints.
Prepare Your Manuscript
Before sending your manuscript to the TTS API, ensure your text is in a clean, plain format. Removing unnecessary formatting like bolding or italics helps the engine focus on the narrative flow. This preparation step is crucial for self-publishers who want consistent audio quality across their entire book.
Clean Text Format
Start with a simple .txt file. The API processes plain text efficiently, so remove complex layout elements from your manuscript. This ensures the speech engine reads your story exactly as intended, without getting confused by hidden characters or strange formatting codes.
- Use plain .txt files
- Remove markdown styling
- Ensure consistent spacing
Remove Unwanted Symbols
Clean up your text by deleting any special symbols that might disrupt the reading flow. Characters like asterisks, hashes, or unusual quotation marks can cause the engine to stumble. A clean text file leads to smoother, more natural-sounding narration for your audiobook.
Check Character Count
Keep each request under 4,096 characters. This limit ensures efficient processing and helps you manage your usage budget effectively. Knowing your character count allows you to split long chapters into manageable segments, making it easier to track costs and generate audio files without errors.
- Max 4096 characters per request
- Track characters for billing
- Split long chapters
Choose Your Voice
Choosing the right voice sets the tone for your audiobook. Our eight presets are crafted from licensed recordings, offering distinct personalities to match your story’s mood without hiring multiple actors.
Select Aria for Narration
Aria brings a warm, clear narrator voice ideal for fiction and memoirs. Her natural pacing helps listeners stay engaged with complex plots, making her a go-to choice for self-published authors seeking a professional yet approachable sound.
- Warm and clear tone
- Ideal for fiction and memoirs
- Natural pacing for engagement
Pick Orion for Deep Tone
Orion offers a deep, documentary-style voice perfect for non-fiction, history, or serious dramas. His authoritative delivery adds weight to factual content, helping you create immersive audio experiences that command attention.
- Deep, documentary-style delivery
- Great for non-fiction and history
- Adds authority to content
Match Tone to Genre
Match your voice to your genre for consistency. Use Nova for upbeat young adult novels or Atlas for neutral news-style readings. Each voice is based on licensed recordings, ensuring high-quality synthetic speech that resonates with your target audience.
- Nova: bright and upbeat
- Atlas: neutral and news-like
- Consistent genre-appropriate tone
Set Language Parameters
Language handling is straightforward: you can explicitly define the output language or rely on automatic detection. For non-English texts using Latin script, setting the language parameter is essential to avoid defaulting to English. For example, use de for German or es for Spanish to ensure accurate pronunciation and intonation.
Specify German Code
When generating audiobooks in German, explicitly pass language: "de" in your request. This ensures the TTS engine applies the correct phonetic rules and prosody for German narration, preventing the voice from sounding like an English speaker with a German accent.
- Use ISO 639-1 code
de - Works with all 8 voices
- Ensures natural German cadence
Use Default English
If you omit the language field and provide text in Latin script, the system defaults to English. This is convenient for English manuscripts but risky for other Latin-based languages like French or Italian, which may be mispronounced as English.
Guess from Script
The API guesses the language based on the script (e.g., Cyrillic vs. Latin). However, explicit specification is recommended for mixed-language texts or less common languages to guarantee accuracy. Unsupported languages like Vietnamese or Thai are not available, so check the list before processing.
- Auto-detects script type
- Explicit is safer than implicit
- Check full 23-language list
Make the API Request
To start narrating your manuscript, you need to send your text to our API endpoint. This process is straightforward and works with any standard HTTP client, allowing you to integrate audiobook generation directly into your self-publishing workflow without complex setup.
Use cURL Command
The quickest way to test the API is using cURL. You can run a simple command in your terminal, replacing YOUR_API_KEY with your unique key from the dashboard. This method requires no additional libraries or code installation.
- Replace YOUR_API_KEY with your actual key
- Set the Content-Type to application/json
- Point to https://api.44bd.info/v1/audio/speech
Send JSON Payload
Your request body must be valid JSON containing the text, voice selection, and desired format. The model parameter accepts either tts-1 or tts-1-hd. Ensure your text stays under 4096 characters per request to avoid truncation.
Receive Audio File
Upon successful processing, the API returns the audio file directly in the response body. You can save this binary data as an MP3, WAV, Opus, FLAC, or PCM file. Failed requests return a full refund automatically.
Handle Long Texts
Your manuscript likely exceeds the 4,096-character limit per request. You must split your text into smaller parts before sending them to the API. Each part generates a separate audio file that you will later combine.
This approach ensures consistent quality and prevents request failures. By processing chunks individually, you maintain control over the narration flow and can handle errors more easily.
Split into Chunks
Divide your text into segments of up to 4,096 characters. Ensure you do not cut mid-sentence to preserve natural pausing. Spaces and punctuation count toward this limit, so plan your breaks carefully.
- Maximum 4,096 characters per request
- Preserve sentence boundaries for natural pauses
- Count spaces and punctuation as characters
Generate Segments
Send each chunk to the API endpoint. The system processes requests sequentially, so longer texts may queue. Each successful request returns an audio file corresponding to that specific segment.
Join Audio Files
Once all segments are generated, concatenate the audio files in order. Use standard audio editing tools to merge the files. This creates a continuous audiobook file ready for publishing.
Choose Output Format
Selecting the right audio format is crucial for your audiobook distribution workflow. Our API supports MP3, WAV, Opus, FLAC, and PCM, allowing you to balance file size and quality based on where your content will live. Each format serves a specific purpose in the publishing pipeline.
Understanding these options ensures your listeners get the best experience without unnecessary data overhead. You can switch formats easily by changing the response_format parameter in your API request.
MP3 for Web
MP3 is the universal standard for web players and most audiobook platforms. It offers a good balance between file size and acceptable audio fidelity, making it ideal for streaming or downloading on mobile devices.
Most self-publishing platforms prefer MP3 because of its widespread compatibility with devices and software.
- Best for general distribution
- Supported by all major audiobook platforms
- Small file size for faster uploads
WAV for High Quality
WAV provides uncompressed, lossless audio. Use this format if you need the highest fidelity for archival purposes or if your target platform requires uncompressed audio for review.
It results in larger files, so consider storage implications when generating many chapters.
- Lossless quality
- Ideal for archival storage
- Larger file sizes
Opus for Compression
Opus offers superior compression efficiency, delivering high quality at lower bitrates. This is excellent for reducing bandwidth costs while maintaining clear narration.
It is increasingly supported by modern streaming services and is great for efficient storage.
- High efficiency
- Low bandwidth usage
- Modern streaming support
Check Your Balance
Track your consumption directly from the dashboard to ensure you stay within budget. The interface displays your remaining balance, trial credit status, and total characters generated, giving you full visibility into your spending without hidden surprises.
View Character Usage
Each request consumes characters based on the input text length, including spaces and punctuation. Your dashboard updates in real time, showing exactly how many characters remain in your current balance or trial period.
- Real-time character counter
- Trial credit tracking
- Usage history log
Monitor Costs
At $0.01 per 1,000 characters, costs are predictable and transparent. Failed requests are free, so you only pay for successfully generated audio. This model ensures you never overpay for errors.
Top Up with USDT
Add funds instantly using USDT on the TRON network (TRC-20). The minimum top-up is $5, and credits appear in your account within minutes. No credit cards, PayPal, or subscriptions are required.
- TRC-20 USDT only
- No minimum amount
- Instant credit

Review the Audio
Before publishing, play the generated audio to verify natural pacing and clarity. Synthetic speech carries an inaudible watermark, but the primary focus is ensuring the narration flows logically without robotic glitches or awkward pauses that might distract listeners.
Listen for Quality
Use a voice like Aria for warm narration or Orion for a deep, documentary style. Listen for consistent tone across chapters. If the audio sounds too fast or slow, adjust the speed parameter between 0.25 and 4.0 in your next request.
- Check for consistent emotional tone
- Verify pacing matches the genre
- Listen for robotic artifacts
Check for Errors
Ensure no text was skipped or garbled during generation. Since inputs are limited to 4096 characters, verify that chunk boundaries did not cut off sentences mid-thought. If an error occurred, the request was free, so you can retry without cost.
- Verify complete text coverage
- Check chunk boundaries
- Confirm no data loss
Download Final File
Save the returned audio file in your preferred format, such as MP3 for web compatibility or WAV for high quality. The audio is not stored on our servers, so you must download and archive it locally immediately after receiving the response.
- Download MP3 or WAV files
- Archive locally immediately
- Verify file integrity
Publish Your Audiobook
Once your audio files are generated and verified, you are ready to bring your audiobook to market. This step involves distributing your finished product to listeners via established digital platforms and community channels.
The API handles the heavy lifting of narration, but the final presentation relies on your metadata and distribution choices. Properly preparing your files ensures a professional listener experience across all devices.
Upload to Platforms
Submit your generated MP3 files to major audiobook distributors. Ensure your audio meets standard bitrates and formats before uploading to maintain quality for listeners.
Most platforms require a specific file structure. Verify that your segments are correctly ordered and named to prevent playback issues during distribution.
- Check platform audio requirements
- Upload segmented MP3 files
- Verify chapter markers
Add Cover Art
Design a compelling cover image that reflects your book’s genre. Since your voice is synthetic, the visual element becomes crucial for branding.
Ensure your cover art meets the resolution and dimension standards of your chosen distribution platforms to avoid rejection or poor display.
- Design genre-specific cover
- Check resolution standards
- Include title and author name
Share with Readers
Promote your new audiobook through social media and author networks. Highlight the multilingual capabilities if you have translated versions using different voices.
Engage with your audience by sharing snippets of your narration. Use the distinct character of voices like Aria or Orion to showcase the production quality.
- Share social media snippets
- Engage with readers
- Highlight multilingual options
Optimize Your Workflow
Once your basic workflow is stable, you can save time by automating the generation process. Instead of manually triggering each chapter, you can write a script that reads your manuscript, splits it into manageable chunks, and sends requests to the API in sequence. This ensures consistent narration across your entire book without requiring constant supervision.
By leveraging the official OpenAI SDK or simple HTTP libraries, you can integrate the TTS API directly into your publishing pipeline. The SDK handles authentication and basic request management, allowing you to focus on content rather than connection management. This approach is ideal for self-publishers who want to generate audiobooks efficiently without hiring additional staff.
Automate Requests
You can use simple scripts to loop through your chapters and submit them one by one. Since the API processes requests sequentially, automating this ensures a steady stream of audio files without overwhelming your key’s rate limit of 60 requests per minute.
Setting up a queue system allows you to batch multiple chapters together. This is particularly useful for longer manuscripts where you want to generate all segments overnight while you sleep, ready for review in the morning.
- Use Python or Node.js scripts to automate chapter-by-chapter submission.
- Implement a queue to manage the 60 requests-per-minute limit effectively.
- Schedule jobs during off-peak hours to ensure fast processing speeds.
Use SDK Libraries
The API is compatible with the official OpenAI SDK, making integration straightforward. You only need to change the base_url to https://api.44bd.info/v1 and use your API key. This allows you to use familiar code structures and methods you may already know from other TTS services.
Using an SDK simplifies handling the JSON payload and response parsing. It reduces boilerplate code and helps manage edge cases like network timeouts or temporary server errors, ensuring your generation pipeline remains robust even when dealing with large volumes of text.
- Switch
base_urlto the API endpoint for easy integration. - Use standard libraries to handle JSON requests and responses.
- Reduce boilerplate code with familiar SDK methods.
Batch Processing
For large projects, processing chapters in batches can significantly reduce total generation time. By sending multiple requests concurrently (within rate limits), you can generate audio for an entire book in a fraction of the time it would take manually.
Batch processing also helps in managing character counts more effectively. You can group chapters by length to optimize the number of requests, ensuring that each batch stays within the 4096-character limit per request while maximizing throughput.
- Send multiple requests (within the 60 requests/minute limit) to speed up generation.
- Group chapters by length to optimize request counts.
- Monitor batch progress to ensure all segments are generated successfully.
Troubleshoot Common Issues
Running into limits while generating your audiobook? Our API processes requests sequentially, so long texts may queue during high load. If you hit the 60 requests per minute cap, wait a moment or split your manuscript into smaller segments to keep generation smooth and predictable.
Text encoding issues can cause silent failures or garbled output. Ensure your manuscript is UTF-8 encoded and free of hidden symbols before sending. A clean text file prevents unexpected errors during synthesis.
Handle Rate Limits
You have 60 requests per minute after your first top-up (10 before). Requests are processed one after another, so under load a request may wait in a queue. Splitting long texts into 4096-character chunks helps manage this flow.
If you exceed the limit, your request will wait until capacity opens up. No data is lost; simply retry after a brief pause or distribute requests over a longer period.
- Limit: 60 requests/minute per key (10 before first top-up)
- Process: Sequential, so long texts may queue
- Solution: Split text into 4096-character chunks
Fix Text Encoding
Ensure your input text is UTF-8 encoded. Hidden symbols or non-standard characters can cause synthesis errors or unexpected audio artifacts.
Use a simple text editor to clean your manuscript before sending it to the API. This ensures consistent character counting and accurate pronunciation.
- Use UTF-8 encoding
- Remove hidden symbols
- Clean text before API call
Resolve Audio Glitches
Check your audio files for clarity. If you hear glitches, verify that the text length is within the 4096-character limit per request.
If issues persist, try adjusting the speed parameter or selecting a different voice preset that better matches your text's tone.
Remember, failed requests are free, so you can experiment without worrying about extra costs.
- Check text length (max 4096 chars)
- Adjust speed parameter (0.25–4.0)
- Try different voice presets
Next Steps for Creators
After mastering the basics, you can expand your catalog by exploring the remaining six voices. Pair Aria’s warm narration with Felix’s energetic promo style to diversify your content types.
Explore More Voices
Once you are comfortable with Aria, try Orion for deep, documentary-style chapters or Luna for soft storytelling. Each voice offers a distinct tone that can enhance different genres without requiring additional production work.
Try Different Speeds
Adjust the speed parameter between 0.25 and 4.0 to fit your pacing preferences. Slower speeds improve clarity for dense technical texts, while faster speeds can make light reading feel more dynamic.
Scale Your Library
As your collection grows, use the API to generate entire books efficiently. With a balance that never expires and top-ups via USDT, you can maintain a steady production pipeline for self-publishing without monthly commitments.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Special attention is paid to consonant clarity and vowel shapes, ensuring that complex words are understood without distortion.
23 languages
Any voice speaks any of these languages.
