Troubleshooting & FAQ

Solutions for common issues with WTN Suite

Troubleshooting & FAQ

Having an issue? Find solutions below, organized by category.


Installation Issues

⚠️ Warning

Windows Defender or antivirus blocks the installer

WTN Suite is packaged with PyInstaller, which some antivirus software flags as a false positive. This is common with many Python-based applications and is not a security risk.

Fix: Add WTN Suite to your antivirus exclusion list, or click "Run anyway" when Windows Defender SmartScreen appears.

⚠️ Warning

"Missing DLL" or "VCRUNTIME" error on startup

Your system may be missing the Visual C++ runtime libraries.

Fix: Download and install the Microsoft Visual C++ Redistributable (x64 version), then restart WTN Suite.


Account & Login

⚠️ Warning

Verification email not arriving

Check your spam/junk folder first. Firebase sends verification emails from noreply@webtoonnarratorapp.firebaseapp.com.

Fix: Click "Resend Verification" in WTN Suite. If it still doesn't arrive, try registering with a different email provider (Gmail works reliably).

⚠️ Warning

"Session expired" or getting logged out

Firebase auth tokens expire after about 1 hour of inactivity.

Fix: Simply log in again. Enable "Remember Me" / Auto-Login in Settings to stay logged in between sessions.


API Key Problems

⚠️ Warning

"Invalid API key" error

Your API key may be incorrect, expired, or for the wrong provider.

Fix:

  1. Go to Settings in WTN Suite
  2. Make sure you've selected the correct provider (Google AI, OpenAI, or Anthropic) in the Active AI Service dropdown
  3. Re-copy your API key — make sure there are no extra spaces
  4. Verify your key is active at Google AI Studio, OpenAI Platform, or Anthropic Console

⚠️ Warning

"Quota exceeded" or "Rate limit" error

You've hit the usage limit for your API plan.

Fix:

  • Google AI free tier: Wait a few minutes — the rate limit resets quickly. You get ~15 requests per minute on the free tier. See Google AI rate limits for full details.
  • OpenAI: Check your usage dashboard. You may need to add credits. See OpenAI rate limits for tier details.
  • Anthropic: Check your usage dashboard. You may need to add credits. See Anthropic rate limits for tier details.
  • Reduce load: Process one chapter at a time. WTN Suite can handle up to 500–700 images per run, but large batches increase the risk of hitting rate limits mid-run and wasting API requests.

💡 Tip

Best practice: Always process one chapter at a time. Even though WTN Suite supports up to 500–700 images per run, keeping batches to a single chapter (~30–80 panels) ensures clean results and avoids wasting API quota if a rate limit interrupts a large run.

ℹ️ Info

How much do API calls cost?

  • Google AI (Gemini): Free tier is generous for most users. Paid tier starts at ~$0.001 per panel. See Google AI pricing.
  • OpenAI (GPT-4V): ~$0.01–0.03 per panel depending on image size. See OpenAI pricing.
  • Anthropic (Claude): ~$0.01–0.03 per panel depending on model. See Anthropic pricing.

A typical 40-panel chapter costs less than $0.50 with Google AI or OpenAI. Anthropic is similar in cost to OpenAI.


Pipeline Issues

⚠️ Warning

Too many or too few panels detected

The panel detection AI may over-split or under-split depending on the art style.

Fix:

  • Try switching the detection model in Settings → Panel Detection Model: use Manhwa Style for vertical scroll webtoons, Manhua Style for page-based manga
  • Click on any page image to open the Crop Editor and manually adjust panel boundaries
  • For complex page layouts, consider using the Manga Cropper tool first

⚠️ Warning

Scene analysis returns vague descriptions

The AI may produce generic descriptions like "a person stands in a room" instead of specific details.

Fix:

  • Make sure your panel images are clear and not too small — Auto Character Handling works better with high-quality images
  • Try switching AI providers (some models perform better on certain art styles)
  • For precise character names, optionally use the Character Builder to assign custom names

⚠️ Warning

Narrative text is too short or too long

The generated narration may not match your desired style.

Fix:

  • Try a different Narrator Prompt Template in Settings (Master Prompt v4.1 works best for most content)
  • Write a Custom Prompt in Settings to control narration style, length, and tone
  • Use the Script Rewriter (Pro plan) to polish the output

Audio Generation

⚠️ Warning

Edge-TTS: "Network error" or no audio generated

Edge-TTS requires an internet connection since it uses Microsoft's cloud voices.

Fix: Check your internet connection. If the issue persists, try again after a few minutes — the service occasionally has brief outages.

⚠️ Warning

Kokoro-TTS: "eSpeak-NG not found" error

Kokoro-TTS needs eSpeak-NG installed for phoneme processing.

Fix: Download and install eSpeak-NG from GitHub. After installation, restart WTN Suite.

⚠️ Warning

Kokoro-TTS: Very slow generation

If Kokoro-TTS is running much slower than expected, it may be using your CPU instead of your graphics card.

Fix:

  • Make sure you have an NVIDIA graphics card with updated drivers
  • Kokoro-TTS runs ~10x faster with a graphics card — without one, expect 2–5x real-time speed per clip
  • Consider using Edge-TTS instead if you don't have a compatible graphics card

Video Export

⚠️ Warning

Video has black frames or missing panels

Some panels may not have matched correctly with their audio files.

Fix: Make sure your image files and audio files are named in the same order (both numbered sequentially). The Video Compositor matches them by filename order.

⚠️ Warning

Audio is out of sync with panels

This can happen if some audio clips are significantly longer than others.

Fix: The Video Compositor automatically adjusts panel duration to match audio length. If you still see sync issues, try exporting individual clips and checking which ones are mismatched.


Subscription & Billing

⚠️ Warning

Feature shows "Upgrade Required" but I'm subscribed

Your subscription may not have synced yet, or your session may be stale.

Fix:

  1. Log out and log back in
  2. Check your subscription status in My Account
  3. If the issue persists, visit your billing portal to verify your subscription is active

ℹ️ Info

How do I cancel or change my plan?

Open the Stripe billing portal from My Account → Manage Billing. You can upgrade, downgrade, or cancel anytime. When upgrading, you get prorated credit (you only pay the difference for the rest of your billing period).


Getting More Help

If your issue isn't covered here:

  • Discord community: Join our Discord — ask questions and get help from other users
  • Report a bug: Describe the issue in our Discord #support channel with screenshots if possible
  • Email: Contact us through the Discord server for account-specific issues