Speech Troubleshooting Guide

This guide covers both Read Aloud (text-to-speech) and Dictation (speech-to-text) features.

---

Read Aloud (Text-to-Speech)

Overview

The Read Aloud feature uses the browser’s built-in Web Speech Synthesis API. The available voices, quality, and behavior are entirely controlled by your browser and operating system — Storytellr has no ability to install voices or override browser limitations.

Known Browser/Platform Limitations

These are confirmed bugs or limitations in the browsers themselves. They cannot be fixed in Storytellr.

Platform	Issue
Edge on Android	getVoices() always returns empty — voice selection dropdown will be blank. This is a confirmed Microsoft bug with no fix as of 2025. Read aloud may still work using the browser’s default voice.
Chrome/Firefox on Android	Voice selection is ignored — the browser always uses its default voice regardless of what is selected. This is a known Chromium bug.
iOS/iPadOS Safari	Only low-quality built-in voices are exposed via the Web Speech API. Enhanced and Premium voices downloaded through iOS Settings are not accessible to web apps. This is an Apple limitation.

Voice Quality Tips

• Windows: Microsoft Natural and Neural voices (e.g. “Microsoft Brian Online”) are high quality but require an internet connection. Install them via Settings → Time & Language → Speech.
• macOS: Enhanced and Premium voices are available. Install them via System Settings → Accessibility → Spoken Content.
• Android: Use Chrome or Firefox. Voice quality depends on the TTS engine installed on your device (Settings → Accessibility → Text-to-speech).
• iOS/iPadOS: Quality is limited by Safari’s API restrictions. No workaround is available.

Voice Dropdown is Empty

If the voice dropdown shows no options:

1. Edge on Android: This is the known Microsoft bug described above. Try Chrome or Firefox instead.
2. Other browsers: Your browser may not support the Speech Synthesis API, or no TTS voices are installed on your device. Check your OS text-to-speech settings.

---

Dictation (Speech-to-Text)

Overview

The speech-to-text (dictation) feature uses the Web Speech Recognition API, which is built into modern browsers (Chrome, Edge, Safari). This API requires an internet connection because it uses Google’s cloud-based speech recognition servers.

Common Issues

Network Error

Symptom: The microphone icon activates but immediately shows a “network error”

Cause: Something is blocking the connection to Google’s speech recognition servers

Solutions:

1. Check Antivirus/Firewall
   • Temporarily disable your antivirus software
   • Check Windows Defender Firewall settings
   • Look for any security software that might block web requests
2. Check Browser Settings
   • Ensure microphone permissions are granted
   • Check that site permissions allow microphone access
   • Try resetting browser permissions
3. Network Configuration
   • Disable VPN if you’re using one
   • Try a different network (e.g., mobile hotspot)
   • Check if you’re on a corporate network with restrictions
   • Contact IT department if on managed network
4. Browser Extensions
   • Disable ad blockers
   • Disable privacy/security extensions temporarily
   • Try in incognito/private mode

No Microphone Permission

Symptom: Browser doesn’t ask for microphone permission

Solution:

1. Go to browser settings
2. Navigate to Site Permissions → Microphone
3. Ensure the site has permission to use microphone

Microphone Not Working

Symptom: Permission granted but no audio detected

Solution:

1. Check Windows Sound Settings
2. Ensure microphone is not muted
3. Test microphone in other apps
4. Check microphone volume level
5. Try selecting a different microphone if multiple are available

Testing Speech Recognition

To test if the issue is with your browser/system configuration:

1. Visit Google’s official demo: https://www.google.com/intl/en/chrome/demos/speech.html
2. If this doesn’t work, the issue is with your system, not our app
3. If it works, there may be a configuration issue with our implementation

Browser Support

• ✅ Google Chrome (HIGHLY RECOMMENDED) - Full support, best performance
• ⚠️ Microsoft Edge - Known network error issues with speech recognition. Use Chrome instead.
• ✅ Safari (macOS) - Generally works well
• ❌ Firefox - Limited/no support for Web Speech API
• ❌ Opera - Limited support

Microsoft Edge Known Issue

Microsoft Edge has a known bug with the Web Speech API that causes immediate “network” errors when attempting to use speech recognition, even though:

• The microphone permission is granted
• The API is supported
• The same code works perfectly in Chrome

Solution: Use Google Chrome for speech-to-text dictation.

Why this happens: This appears to be an Edge-specific bug in how the browser communicates with Google’s speech recognition servers. Even though Edge and Chrome are both Chromium-based, they have different implementations of certain APIs.

Still Having Issues?

If you’ve tried all the above and still can’t get it working:

1. Check the browser console (F12) for detailed error messages
2. Look for ERR_BLOCKED_BY_CLIENT or similar network errors
3. Check the Network tab in DevTools for failed requests to Google servers
4. File an issue on GitHub with console logs and system details

Technical Details

The Web Speech API makes requests to:

• www.google.com/speech-api/
• speech.googleapis.com

If these domains are blocked by your network or security software, speech recognition will not work.