Frequently Asked Questions

Refund requests are handled case-by-case. If you believe you qualify for a refund (for example, a duplicate or accidental purchase), please email hello@authorvoices.ai with your order/transaction ID, the email used for the purchase, and a short explanation. We'll review your transaction and whether any credits were used; approved refunds are processed via Stripe and typically appear on your statement within 5–7 business days. For complex disputes or suspected fraud, we may follow up for additional verification.

Your current credit balance and unified purchase/usage history are available on the Credits page (accessible from your dashboard). Purchases are applied immediately to your account balance; check the Transactions table on that page for details on purchases, credits used per section, and running available balance.

AuthorVoices.ai is an AI-powered audiobook narration platform that helps independent authors create professional-quality audiobooks using curated AI voices.

Simply upload your EPUB or text file, select a narrator voice, and our system will generate your audiobook in chapters. You can edit individual sections and regenerate as needed.

Cloned voices created from your uploaded sample are private to your account. If you create a cloned voice before buying credits, we retain it for a short trial window (approximately 24 hours) so you can experiment. To keep the cloned voice longer-term, purchase any credits in your account — cloned voices are preserved while you have credits. If you delete a cloned voice it cannot be recovered; you can upload a new sample to recreate one.

Yes — you can try voice previews without purchasing credits. The site provides a limited number of free voice-preview attempts so you can hear narrators and pick the right voice before buying credits. If you run into limits or need more previews for testing, contact support.

Some older/imported projects are preserved as "Download Only" — this means the project files and final audio remain available for download, but the project is not editable in the new editor. Editable projects can be updated and re-narrated as usual.

Processed output files are available for download for approximately 6 hours after the job finishes. After that they are automatically deleted from the server. Download your files as soon as processing is complete — they cannot be recovered once the retention window expires.

The fastest way to reach us is hello@authorvoices.ai or the Contact page on the site. When you contact support, include your account email, any relevant project or section IDs, and screenshots or error messages to help us investigate more quickly.

Cloned voices are private to your account and will appear in your narrator list when you create or edit projects/sections. Select your cloned voice as the default narrator for a project or override the narrator for individual sections when editing. Note: cloned voices created during a trial window are preserved while you have credits (see the Clone Your Voice page for details).

Go to Voice Clone (in your dashboard) and upload a recording of your voice — supported formats are MP3, WAV, M4A, MP4, AAC, OGG, and WEBM. At least 30 seconds of clear speech gives the best results. The system trims and processes the sample, creates a private voice on ElevenLabs, and immediately adds it to your narrator list. You can replace your cloned voice at any time by uploading a new sample. Note: cloned voices are private to your account and are preserved while you hold credits. Free accounts may upload up to 3 voice samples per day; paying accounts may upload up to 10.

Yes — new accounts receive a small registration bonus (5,000 characters, roughly 5–6 minutes of finished audio) so you can test narration quality before purchasing a credit pack. You can also preview any narrator voice on the homepage without using account credits. To produce a full chapter or book you will need to purchase a credit pack from the Pricing or Credits page.

Yes — use the Quick Fix (advanced selection) feature in the section editor. Highlight the specific text passage you want to re-record, choose an alternate phrasing or regenerate with a different voice setting, and only that passage is re-synthesised and stitched back into the section audio. This saves credits and is much faster than regenerating an entire chapter.

Yes — if you have an account on SelfPublishing.pro, you qualify for a 10% AuthorPass discount on credit pack purchases. The discount is applied automatically at checkout once your cross-site membership is verified. If you believe you qualify but are not seeing the discount, contact support with both account emails so we can verify and apply it manually.

Yes — open your project and use the Cover Art option to upload a square image (JPG or PNG). The cover is embedded in the M4B export so it appears as album art in audiobook players. It is also used as chapter art in the ZIP export. For best results, use a square image at least 1400×1400 pixels. You can replace the cover at any time; re-export the project to apply the updated art.

Yes. On any project page you'll see a "Narrate Whole Book" button. Click it and AuthorVoices shows a pre-flight summary — how many sections will be narrated, exactly how many credits it will cost, and your current balance — then kicks off the whole book in the background. You can leave the page and come back; narration keeps running. When you return, each section will show its live status and the top card will show overall progress.

Yes — on the project page, drag and drop sections into the order you want. The new order is saved immediately. Reordering does not affect the audio files already generated; it only changes how sections are sequenced when you export.

Marking a section as Proofed is a personal quality-review flag — it simply means you have listened to that section and are happy with it. It does not affect generation, downloads, or billing. The project page shows how many sections are proofed so you can track your review progress across a long book.

Yes — we offer a free tier so you can explore the platform. The free tier includes a limited amount of narration credits and free voice-preview attempts so you can try voices and the editor. No credit card is required to start; when you’re ready to produce more audio, you can purchase one-time credit packs from the Credits/Pricing page.

You can purchase one-time credit packs from your dashboard (Credits) or the Pricing page. There are multiple credit tiers to match different project lengths. The purchase table displays both the regular price and any current sale price — the charged amount is what appears on your statement. Purchases are applied to your account balance immediately and (under normal policy) credits do not expire. For account-level upgrades, subscription questions, or help with purchases, contact support.

We use a character-based credit model. One hour of finished audiobook audio corresponds to about 9,300 words or roughly 50,000 characters, and credit packs are sized accordingly. Use the Pricing/Credits page to estimate how many credits your manuscript will need.

Narration is sold as one-time credit packs sized by finished audiobook length. One hour of finished audio is roughly 9,300 words (≈50,000 characters), and our credit packs are sized accordingly. The site displays both the regular price and any active sale price for each pack; the exact charged amount is shown at checkout and is what appears on your statement. Credit tiers and any promotional sale prices are configurable (so prices may vary over time). Purchased credits are applied to your account immediately and, under normal policy, do not expire — check your Credits page for current balance before buying more.

You can upload EPUB and Microsoft Word (.docx) files to create a project and auto-detect chapters. If you prefer, you can create a project manually by pasting text. The server accepts large files (up to 2 GB), though typical manuscripts are well within this limit. If you hit an upload error, try splitting the file or contact support with the file type, size, and any error message shown.

Yes. Once your audiobook sections are generated you can download the audio. From the project page you can export a ZIP of individual MP3 files (one per chapter) or a single M4B audiobook file with embedded chapter markers and cover art. Audio generated on AuthorVoices is mastered to ACX-compatible levels, so it is ready for most audiobook retailers. For MP3s recorded outside AuthorVoices, use the Distribution Ready Tool to convert them to retailer specs.

Yes. The system is built around section-level narration so you can update specific parts of a project without having to rebuild the entire audiobook every time. That makes revisions faster and more practical.

The Distribution Ready Tool converts single MP3 files or ZIPs of MP3s into retailer-ready delivery packages (ACX-style). Upload your file and the tool processes each track to recommended specs, shows live progress, and produces a finished download. Guests can process 1 file per 24-hour period (up to 100 MB per file); registered users can process 3 files per 24 hours. Paid users and admins have higher upload limits (up to 2 GB per file). Audiobooks generated directly with AuthorVoices are already exported in distribution-friendly formats, so this tool is mainly for outside MP3 files you need to bring up to retailer specs. Processing credit packs are available for high-volume use.

Manuscript uploads (EPUB / DOCX) are subject to the Flask app's MAX_CONTENT_LENGTH, which is set to 2 GB in app.py (overriding the 50 MB value in config.py). There is no separate route-level byte limit on the manuscript upload endpoint. In practice, advise users to split very large manuscripts or send the file to support. The Distribution Ready Tool accepts MP3 or ZIP uploads with separate limits: 100 MB for free/guest users and up to 2 GB for paid users and admins. These limits are enforced at the route level. Guidance for agents: - If a user cannot upload an EPUB or DOCX, confirm the exact error message, file size, and file type before escalating. - If the user is uploading to the Distribution Ready Tool and hits an error, check user account status (free vs paid) and confirm they used the correct endpoint before escalating. - When in doubt, capture error message, account email, file size, file type, and the route used; escalate to Engineering with these details. Where to check: for Distribution Ready behavior, review the distribution_ready template and the corresponding upload handler in app.py. Action: Use this FAQ when responding to upload-size questions or triaging upload errors; update if server behavior changes.

Generation time varies by section length and current queue. Short sections often synthesize in minutes; full books or many sections can take hours. You can monitor live progress on the project/section pages (and the Distribution Ready Tool shows per-file status). If you see unusually long delays, contact support with your project details.

Credits are charged per section, not up front, and only right before that section actually renders. That means: (1) if a section errors, you are not charged for it; (2) if you run out of credits part-way through, the batch stops cleanly at the section it could not afford — already-narrated sections stay; (3) if you click "Stop After Current Section", AuthorVoices finishes the section in flight and then halts — no half-finished audio files. By default, the batch skips sections that already have audio, so you will not be re-billed for work you have already paid for. If you've edited the source text of an already-narrated section and want it re-rendered, tick the "Also re-narrate sections that are already rendered" checkbox in the pre-flight modal before starting.