Voice and multimodal AI

About this course

This course covers using voice, camera, and file uploads with AI instead of typing everything into a text chat. It addresses a real gap: many everyday tasks — photographing a receipt, dictating a thought on the go, parsing a contract — still get done by hand because people don't realize AI can handle them through voice, photos, or documents.

The course moves through input channels one by one: dictation and voice conversation first, including transcribing meetings and lectures, then camera-based work — asking questions about photos, reading screenshots, and recognizing text in images — followed by uploading documents, reviewing contracts and tables, and checking citations against a "not in the document" rule. A separate block covers hands-free scenarios on the go, at home, and in transit.

The course suits anyone already comfortable with a text-based AI chat who wants to extend that to voice, photos, and documents for everyday and work tasks. No technical background is needed — just a smartphone or computer with a camera and microphone, and willingness to try new ways of talking to AI.

What you'll learn

• Dictate requests so AI understands them the first time
• Hold a voice conversation with an assistant about a real task
• Transcribe meetings, lectures, and notes into text
• Ask questions about photos and recognize where the model gets it wrong
• Read screenshots and recognize text embedded in images
• Upload documents and get summaries checked against citations
• Review contracts and tables through file uploads
• Choose the right channel — voice, photo, or text — for a given situation

What you'll end up with

• A working set of voice and visual scenarios tested on real tasks
• The habit of turning to AI via photo or voice in situations you used to type out
• One important personal document fully broken down, with a content map and checked answers
• A personal reference for which channel to use when — voice, photo, or text

This course isn't for someone who hasn't tried a text-based AI chat at all yet — start with a basic prompting course first, then come back to voice and photo input.