Frequently asked questions
What is the best way to extract text from a PDF?
Extract plain text using any standard PDF reader or library, then paste it into your API request. The API accepts raw text strings, but complex formatting like headers or footers may disrupt the narrative flow. Just ensure the text is under 4096 characters per request to avoid truncation during synthesis.
Does the API handle PDF formatting?
No, the API processes plain text input only. It does not interpret PDF structures like tables, images, or specific font styles. You must extract the raw text content before sending it to the endpoint. Formatting decisions, such as adding pauses, are handled by your application logic, not the API.
How do I split a 500-page book?
Split your text into chunks of up to 4096 characters. Send each chunk as a separate POST request to the speech endpoint. Your application should then join the resulting audio files sequentially. This approach ensures each segment synthesizes correctly without exceeding the single-request character limit.
What is the maximum character limit per request?
The maximum input length is 4096 characters per request. Spaces and punctuation count toward this limit. If your text exceeds this, you must split it into smaller parts. Each part generates a separate audio file that you can concatenate later to form the complete audiobook.
Can I pause and resume generation?
The API does not support streaming or partial audio generation. Each request generates a complete audio file. You cannot pause a single synthesis job. To manage long texts, split them into chunks beforehand. If a request fails, you can retry that specific chunk without affecting others.
How do I ensure voice consistency across chapters?
Specify the same voice parameter in every API request. Choose one of the eight presets, such as Aria or Felix, and use its name consistently. Since voices are static presets, selecting the same voice ensures uniform narration tone throughout your entire audiobook project.
Do I need to install any software?
No software installation is required. You can use the API via standard HTTP requests or the official OpenAI SDK by adjusting the base URL. Any programming language that supports JSON and HTTP can interact with the endpoint. Just ensure you have a valid API key for authentication.
Can I use this on a Mac or Windows?
Yes, the API works on any operating system that supports HTTP requests. Since it is a cloud-based service, your OS does not matter. You can use Python, JavaScript, or any other language on Mac, Windows, or Linux to send text and receive audio files.
What happens if the internet cuts out?
If the connection drops, your request may fail or time out. Failed requests are free, so you will not be charged for incomplete generations. Simply retry the specific text chunk that did not complete. Your API key remains valid, and your balance is unaffected by transient network issues.
How do I debug API errors?
Check the HTTP status code and response body for error details. Common issues include invalid API keys, exceeding rate limits, or invalid voice names. Ensure your text is under 4096 characters and your language code is correct if specified. The API returns clear error messages to help you fix the request.
Is the generated audio royalty-free?
The generated audio files are yours to use. The API does not claim ownership over the output. You can use the audio for personal or commercial projects without paying additional royalties. The voices are based on licensed recordings (LibriTTS-R, CC BY 4.0), and the output is synthetic and ready for your distribution needs.
Can I sell the audiobooks commercially?
Yes, you can sell the generated audiobooks commercially. There are no restrictions on using the output for profit. Since you pay per character used, your costs are predictable. You retain the rights to the audio files generated by your requests, allowing you to distribute them on any platform.
Do I need to credit the API?
Attribution is not strictly required by the API terms, though acknowledging the tool is appreciated. However, the audio files contain an inaudible watermark marking them as synthetic speech. You are free to distribute the audio without explicit credit, though acknowledging the tool is always appreciated by the community of indie authors and publishers.
How do I handle special characters?
The API accepts standard Unicode text, including special characters and punctuation. Spaces and punctuation count toward the character limit. If you use non-English scripts, ensure you set the language parameter correctly. The system guesses the language from the script if not specified, but explicit setting is more reliable.
Understanding the Process
This guide walks indie authors and self-publishers through converting text into audiobooks using the AudioPress API.
By automating narration, you preserve creative control while leveraging a reliable, character-based service designed for long-form content.
Extract Text from PDF
Use a standard PDF reader or library to extract raw text from your manuscript. Ensure formatting artifacts like page numbers or headers are removed to keep the input clean.
Prepare Input Data
Verify that your text does not exceed 4,096 characters per request. Clean up special characters and ensure proper spacing, as spaces and punctuation count toward your character total.
Call the API
Send a POST request to https://api.playv9999.com/v1/audio/speech with your text, selected voice, and desired format. The API returns an audio file ready for your audiobook project.
Technical Requirements
To get started, navigate to the /get-api-key/ page and sign in with your Google account or create a new account using an email and password. This generates a unique Bearer token that authenticates your requests to the API.
API Key Setup
Your authentication header must follow the format Authorization: Bearer tts-.... You can regenerate this key at any time via your account dashboard; the previous key becomes invalid immediately upon issuance, ensuring secure access for your publishing workflow.
- Generate key on /get-api-key/ via Google or email/password
- Use
Authorization: Bearer tts-...header for requests - New keys instantly invalidate the old one
Character Limits
Each API request accepts input text up to 4096 characters. Spaces and punctuation count toward this limit. For longer manuscripts, you will need to split your content into smaller chunks before sending them to the endpoint.
- Maximum 4096 characters per request
- Spaces and punctuation count as characters
- Split long texts into manageable chunks
Network Connectivity
A stable internet connection is required to send text payloads and receive the resulting audio files. Requests are processed sequentially, so under high load, synthesis may queue slightly, but connectivity ensures your publishing pipeline moves smoothly.
Step-by-Step Conversion
Convert your extracted text into audiobooks using the OpenAI Python SDK (or any HTTP client). Install the library via pip, then configure the client with your API key and the base URL https://api.playv9999.com/v1. This setup ensures compatibility with our tts-1 and tts-1-hd models.
Install Dependencies
Use pip install openai to get the necessary tools. No custom SDK is required; the standard library handles requests seamlessly.
- pip install openai
- Python 3.8+ recommended
Write the Script
Import the OpenAI client and initialize it with your tts-... key. Define your text input, ensuring it stays under 4096 characters per request. Select a voice like 'aria' for narration and specify a format such as mp3.
- Set
base_urltohttps://api.playv9999.com/v1 - Use
model='tts-1'ortts-1-hd - Define
voiceparameter (e.g., 'aria')
Run the Conversion
Execute the script to send the POST request. The API returns the audio file directly. For longer books, loop through your text chunks, saving each segment. Finally, concatenate the files to create a complete audiobook chapter.
Handling Long Texts
Long texts must be split into parts of up to 4096 characters before sending them to the API. This ensures each request stays within limits and processes efficiently, as texts are synthesized roughly at real-time speed. You can split the content programmatically and join the resulting audio files seamlessly.
The API does not embed custom metadata. Your local workflow should tag the final audiobook file correctly so distributors and listeners can navigate the long-form content seamlessly. This keeps your production pipeline clean and professional without relying on the API for structural tags.
Subsections
H3: Chunking Strategy
Split your manuscript into segments of 4096 characters or less. Since requests are processed sequentially, keeping chunks manageable helps avoid queue delays. Send each chunk as a separate POST request to https://api.playv9999.com/v1/audio/speech.
H3: Joining Audio Files
Receive individual audio files for each chunk. Use a local tool to concatenate them in order. This ensures the final audiobook flows naturally, preserving the narrative arc without gaps or overlaps.
H3: Metadata Preservation
The API does not embed custom metadata. Your local workflow should tag the final audiobook file correctly so distributors and listeners can navigate the long-form content seamlessly.
Image Prompt
A clean, modern workspace featuring a laptop displaying code for splitting text, next to a stack of neatly arranged audiobook covers and a pair of headphones, symbolizing the technical yet creative process of self-publishing.
Bullets
['Split text into 4096-character chunks', 'Send parallel requests for speed', 'Concatenate audio files locally', 'Tag final file with chapter metadata', 'Use standard formats like MP3', 'Validate audio continuity manually']
Voice Selection Strategies
Choosing the right voice is crucial for maintaining listener engagement across a long-form audiobook. AudioPress provides eight distinct presets. Selecting a voice that aligns with your genre ensures the audio feels authentic rather than robotic.
Consistency is key for serialized content. By locking in a specific voice for all chapters, you create a cohesive listening experience. You can test different presets to find the perfect match for your story's tone before committing to full narration.
Matching Tone to Genre
Match your voice to the narrative style. For mystery or thriller, consider the deep, documentary tone of Orion or the calm instruction of Sage. For light romance or young adult, the bright, upbeat Nova or warm Aria might suit better.
Use Felix for energetic promotional excerpts or Milo for conversational memoirs. Atlas offers a neutral news-style delivery suitable for non-fiction or historical accounts. Ensure the voice complements the text’s emotional weight.
- Orion: deep, documentary style
- Nova: bright, upbeat
- Aria: warm, clear narrator
- Sage: calm, instructional
- Milo: friendly, conversational
- Felix: energetic, promo
Consistency Across Chapters
Stick to one voice for the entire book to maintain continuity. Switching voices mid-book can disorient listeners. Use the same voice preset for all requests to ensure consistent audio characteristics.
If using multiple voices for different characters, document your choices. Keep a style guide noting which voice represents which character type. This helps maintain consistency during production and editing.
Testing Different Voices
Generate short samples for each chapter or section. Use the free trial credit ($0.50) to test up to 50,000 characters across different voices. Evaluate clarity, pacing, and emotional fit.
Adjust the speed parameter (0.25–4.0) if a voice feels too fast or slow. Listen for natural flow and ensure the voice doesn’t become fatiguing over long listening sessions.
- Generate 50-100 character samples
- Adjust speed between 0.25–4.0
- Listen for fatigue factors
- Document voice-character mappings
Post-Processing
After generating the audio files, you often need to standardize formats for distribution platforms like ACX or Kindle Direct Publishing. While our API returns high-quality MP3 or WAV files by default, you may need to convert them to specific specifications required by retailers.
File Format Conversion
Our API supports MP3, WAV, OPUS, FLAC, and PCM outputs. MP3 is ideal for web distribution, while WAV is preferred for professional audiobook masters. Use standard tools like FFmpeg to convert between formats if your target platform has strict requirements.
- MP3: Best for web streaming and general use.
- WAV: Ideal for high-quality audiobook masters.
- FLAC: Lossless compression for archival purposes.
Audio Quality Check
Listen to the generated files to ensure natural pacing and correct pronunciation. Since our API handles up to 4096 characters per request, verify that chunk boundaries do not create awkward pauses. Adjust the speed parameter if the narration feels too fast or slow for the genre.
- Check for consistent tone across chapters.
- Verify punctuation does not cause unwanted pauses.
- Ensure voice matches the story’s mood.
Metadata Embedding
Embed metadata such as title, author, and narrator into the final audio files. This ensures your audiobook displays correctly on platforms. Our API does not handle metadata embedding directly, so use dedicated tools like id3v2 or Audacity to add this information before distribution.
- Add title, author, and narrator info.
- Use tools like id3v2 or Audacity.
- Ensure compatibility with distribution platforms.
Common Pitfalls
Converting PDF text directly often introduces formatting artifacts that confuse the speech engine, leading to awkward pauses or mispronounced characters. Ensuring clean, plain-text input before sending it to the API prevents these audio glitches and maintains narrative flow for your audiobooks.
Even with clean text, hitting limits is common. Publishers must manage character counts and API rate limits to ensure smooth production pipelines without unexpected interruptions or extra costs during bulk conversion jobs.
Punctuation Issues
PDFs often contain line breaks, spaces, or special characters that do not translate well to speech. These artifacts can cause the TTS engine to insert unwanted pauses or stumble over non-standard symbols. Cleaning the text to remove unnecessary whitespace and standardizing punctuation ensures a natural listening experience.
Ensure your preprocessing script strips out page numbers, headers, and footers before submitting the text. This reduces character count and eliminates noise that might otherwise be read aloud as gibberish or strange sounds.
- Remove non-speech elements like page numbers and headers.
- Standardize ellipses and hyphens for consistent pacing.
- Strip invisible Unicode characters that cause parsing errors.
Character Count Exceeded
The API accepts a maximum of 4096 characters per request. If your chapter exceeds this limit, the request will fail. You must split your text into smaller chunks manually or via script before sending.
Each chunk is processed independently, so ensure your splitting logic respects sentence boundaries to avoid cutting off mid-sentence. This preserves the natural flow of the narration when the audio files are joined later.
- Limit each request to 4096 characters.
- Split at sentence boundaries to preserve flow.
- Failed requests are free, but valid ones incur charges.
Rate Limit Hits
Your API key is limited to 60 requests per minute. During high-volume production, hitting this limit can cause requests to queue or fail temporarily. Monitor your usage and implement exponential backoff in your script to handle rate limits gracefully.
For large books, consider spacing out your requests or processing chapters sequentially rather than in parallel bursts. This ensures steady progress without exceeding the 60 requests per minute threshold, keeping your production timeline predictable.
- Limit: 60 requests per minute per key.
- Implement exponential backoff in your script.
- Process chapters sequentially to avoid bursts.

Cost Estimation
Pricing is straightforward: $10 per million characters, or $0.01 per 1,000 characters. Spaces and punctuation count toward the total. Failed requests are free, so you only pay for successful synthesis.
Calculating Character Count
Before calling the API, count your text characters precisely. Since spaces count, a 40,000-character novel excerpt costs $0.40. Use standard character counters to estimate your budget accurately before processing.
- Include spaces and punctuation in your count
- Use a simple text editor to verify length
Budgeting for Large Books
For a 50,000-word novel, expect roughly 250,000 characters. This translates to a $2.50 production cost. With the free trial credit of $0.50, you can test a chapter immediately after signing in with Google.
Optimizing Costs
Remove unnecessary whitespace before sending text to the API. Since each character costs money, cleaning your PDF extraction reduces your bill. Top up with USDT (TRC-20) for instant, fee-free credits that never expire.
Best Practices for Publishers
For indie authors, consistency is key. Automate your workflow by scripting character extraction and API calls, ensuring every chapter follows the same narration rules. This reduces manual errors and lets you focus on editing rather than technical setup.
Automate Your Workflow
Integrate PDF parsing directly into your publishing pipeline. Use the API’s JSON structure to send text chunks automatically, keeping your production steady without constant manual intervention.
Quality Control Steps
Listen to the first 30 seconds of each chapter to verify tone and pacing. Since voices are based on licensed recordings, ensure the selected voice matches the book’s mood before committing to the full run.
- Check for natural pauses in long sentences
- Verify language detection for mixed-script texts
- Confirm character count accuracy before bulk generation
Scaling Production
With a limit of 60 requests per minute, plan your generation queue carefully. Use the 4096-character chunk limit efficiently to maximize throughput while staying within rate limits.
Troubleshooting
Even with robust scripts, audio glitches like silence or abrupt cuts can occur. These usually stem from input text exceeding the 4,096-character limit per request or network interruptions during file retrieval. Always validate text length before sending requests.
Audio Glitches
Check that your chunking logic correctly splits text without breaking words. If silence appears, verify the join process preserves audio boundaries. Failed requests are free, so retrying is cost-effective. (Note: Requests are processed sequentially, so under load a request may wait in a queue.)
- Validate character counts before each API call
- Ensure chunk boundaries do not split sentences awkwardly
- Retry failed requests to recover any lost charges
Language Detection Errors
Default language guessing may misidentify non-English Latin text. Always set the language parameter explicitly for accuracy. This prevents mispronunciation in languages like Dutch or Greek when using default English assumptions.
- Explicitly define ISO language codes in requests
- Avoid relying on script detection for mixed languages
- Test edge cases with ambiguous text samples
API Timeouts
Requests process sequentially under load, potentially causing queue delays. If a timeout occurs, your text remains uncharged. Ensure your script handles HTTP 429 errors by implementing exponential backoff to respect the 60 requests-per-minute limit.
- Implement exponential backoff for retry logic
- Monitor rate limits to stay under 60 req/min
- Check network stability for consistent connectivity
Future Integration
Integrating the TTS API into your CI/CD pipeline allows you to automate audiobook generation alongside your book publishing workflow. By embedding API calls in your build scripts, you can trigger audio synthesis whenever a new manuscript is committed or updated, ensuring your audio assets are always in sync with your text versions.
This automation reduces manual overhead and enables rapid iteration. You can configure your pipeline to handle the chunking and joining logic, delivering ready-to-publish audio files directly to your staging environment.
CI/CD Pipelines
Embed the API call directly into your build scripts (e.g., GitHub Actions, GitLab CI). Use the OpenAI SDK with a custom base URL to generate audio during the build phase. This ensures every code commit can trigger an audiobook preview.
Automate the splitting of long texts into 4096-character chunks within the pipeline. Process requests sequentially or in controlled batches to respect the 60 requests per minute rate limit, then reassemble the audio segments.
- Use environment variables to store your API key securely.
- Set up webhooks to notify you when synthesis is complete.
- Version your generated audio files alongside your source text.
Cloud Storage Integration
After synthesis, automatically upload the generated audio files to cloud storage services like AWS S3 or Google Cloud Storage. This creates a centralized repository for all your audiobook assets, making them easily accessible for distribution or backup.
Store metadata such as character count, voice used, and generation timestamp alongside the audio files. This information is useful for tracking costs and managing library organization.
- Use signed URLs for secure access.
- Automate cleanup of temporary chunk files.
Distribution Platforms
Directly push your generated audiobooks to platforms like Audible, iTunes, or Spotify through their respective APIs or upload interfaces. By integrating the TTS API with your distribution workflow, you can reduce the time from manuscript to market.
Ensure that the audio format (mp3, wav, opus, flac, pcm) meets the specific requirements of each distribution platform. The API supports multiple formats, allowing you to optimize files for different platforms without re-processing.
Maintain consistency in voice selection across all chapters and books in your series. Use the API's voice presets to ensure a uniform listening experience for your audience.
- Check platform-specific format requirements.
- Automate metadata upload with each book release.
Support and Resources
Access comprehensive guides and API references to streamline your audiobook production workflow. Our documentation covers integration details, voice presets, and pricing models to help you get started quickly.
Documentation Links
Explore detailed endpoints, code examples, and troubleshooting tips. All resources are updated to reflect current API behaviors and limits.
- API Reference Guide
- Quick Start Tutorial
- Voice Preset Details
Community Forums
Connect with other self-publishers and developers. Share tips on chunking strategies and voice selection for different genres.
Email Support
Reach out directly for specific questions about your account or integration. We respond via email only, ensuring focused assistance without phone queues.
Support is available for technical issues and billing inquiries. Please include your API key and error logs for faster resolution.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Language support and format quality directly impact accessibility.
23 languages
Any voice speaks any of these languages.
