Voice Mode
Voice mode lets you have a spoken conversation with AI Chat. Ask questions out loud, get audio responses, and study without touching the keyboard.
Starting Voice Mode
Click the microphone button in the chat input to start a voice conversation. The first time you start voice mode, Scholarly briefly explains what's about to happen before your browser asks for microphone access — so you know to click Allow and don't get a surprise system prompt.
Once active, a voice bar appears at the bottom of the screen with controls for the conversation.
Microphone Permission
If your browser blocks the microphone, the voice bar shows the exact unblock steps for your device:
- Desktop browsers — click the lock icon next to the URL and switch the microphone permission to Allow.
- iPhone / iPad — open Settings → Safari → Microphone (or Chrome → Microphone) and allow scholarly.so.
- Android Chrome / Firefox — tap the lock icon in the address bar → Permissions → Microphone → Allow.
Auto-Reconnect
A brief network blip — switching from Wi-Fi to cellular, walking past a dead spot — no longer ends a voice session. Voice mode automatically reconnects up to five times with progressive backoff. If the connection can't be restored, the voice bar shows a clear retry option.
AI Chat Voices
Two AI Chat voices are available:
- Luther — Warm, storytelling style. Good for conceptual explanations, narrative walkthroughs, and making complex topics feel approachable.
- Flora — Calm, logical style. Good for structured breakdowns, step-by-step reasoning, and technical subjects.
Choose between a Male or Female voice. Switch voices mid-session directly from the voice bar, or set a default in Settings.
Voice Quality
Voice mode uses a fast, natural-sounding voice model with strong handling of accents and background noise. Responses feel conversational with natural turn-taking — the AI knows when you have finished speaking and responds without awkward pauses. The model also handles tool use reliably, so you can ask it to look something up in your PDFs or search the web mid-conversation and get a spoken answer back.
Playback Speed
Control how fast the AI speaks with the playback speed selector on the voice bar:
- 1x — Normal speed.
- 1.25x — Slightly faster, good for review.
- 1.5x — Quick pace for familiar material.
- 2x — Double speed for rapid review.
The AI voice stays natural at all speeds — pitch is preserved so it does not sound distorted.
Switching Between Voice and Text
Voice and text share the same conversation. You can switch between typing and talking at any point without losing context.
If you type a message while voice mode is active, the AI responds with audio instead of a regular text reply. This makes it easy to clarify something in writing and still get a spoken answer.
What Voice Mode Can Do
While in voice mode, the AI has the same capabilities as text chat:
- Search your PDFs, notes, and other linked content.
- Search the web for current information.
- Generate study materials like flashcards, podcasts, and video lectures.
- Answer follow-up questions with full conversation context.
- Pull specific information from your uploaded files when you ask about them.
Text Selection and Speak
Outside of voice mode, you can have the AI read any text aloud. Select text in a chat message and click Speak from the floating menu.
Daily Limit
Free users get 15 minutes of voice chat per day; premium users get 60 minutes. The timer resets every 24 hours, and when you run out you will see an upgrade prompt.
Tips
- Use voice mode during commutes or workouts to review material hands-free.
- Start at 1x speed for new topics, then bump up to 1.5x or 2x when reviewing material you already know.
- Combine voice mode with @ content linking — say "explain chapter 3 of my biology PDF" and the AI reads and discusses it with you.