Listening to AI’s Capabilities

Curriculum Integration Statement / Values Statement

AI is becoming increasingly integrated into today’s media landscape. Helping students explore new AI tools for storytelling, design, and production will educate and prepare them for their future. When used thoughtfully, AI can speed up brainstorming, support research, and open the door to experimentation with visuals, audio, and interactive content. The goal isn’t to replace creative skill, but to give students another set of tools to enhance their ideas and prepare them for a media landscape where AI is already part of the workflow.

Setting the Context / Curriculum Integration Goals

This assignment will help provide a specific example of integrating AI into the curriculum, with a large Human Skills component incorporated inside. Partnered students will have the ability to work together practically within their field, and to critically think of the limitless possibilities that AI can assist with. Having multiple pillars (Human Skills and Artificial Intelligence) naturally integrated into a single assignment will help further deepen Seneca’s Curriculum Integration goals.

Assessment Details

Title: Listening to AI’s Capabilities

Deliverables/Objectives:

• Partnered students write & record a 15-30 second piece of dialogue with Seneca’s audio equipment (or Seneca Sandbox studios) in 3 different types of audio quality (good, medium, & poor).
• Students will use their own post-production editing skills to edit each of their audio recordings to reach a higher quality of audio (can use Seneca Sandbox Workstations).
• Students will then use AI audio editing tools (such as Adobe Podcast) to edit their audio recordings.
• Students will reflect on the processes of audio editing, both manually and AI-assisted, and are encouraged to critically evaluate AI’s position in audio processes.

Learning Objectives:

• Students will understand the technical capabilities and limitations of AI-driven audio restoration in film, music, and media.
• Students will develop hands-on skills in audio recording using professional equipment and in audio editing using both manual and AI-assisted methods.
• Students will practice critical listening and analytical comparison between human and AI sound mixing results.
• Students will collaborate in pairs, enhancing teamwork and communication skills relevant to media production environments.

Materials Required:

• Boom microphone and lavalier microphone (Seneca’s audio equipment) for high-quality and varied audio capture — can use the Sandbox Studio spaces as well.
• Audio editing software (e.g., Adobe Audition, Audacity, Pro Tools) installed on computers for manual sound mixing — can use Sandbox Studio workstations.
• Access to AI audio enhancement tools, such as Adobe Podcast - Enhance Speech or similar plugins.
• Headphones for critical listening and comparison of audio outputs.

Preamble: Introduction to Audio Restoration and AI in Media

To begin this assignment, instructors should open with a class discussion on the critical role high‑quality audio plays in film, music, and media, emphasizing how issues such as background noise or poor clarity can significantly diminish a viewer’s or listener’s experience. Introduce foundational concepts including audio restoration, noise reduction, and the growing influence of artificial intelligence in modern post‑production workflows. To ground these ideas, present short before‑and‑after examples of audio that have been restored through both manual editing and AI‑assisted enhancement, encouraging students to actively listen for differences in quality, clarity, and tonal balance. After this demonstration, explain and assign the assignment provided below:

Part 1: Scriptwriting and Planning Recording Sessions

• Students pair up and collaboratively write a 15-30 second dialogue clip, ensuring the script is clear and suitable for recording.
• Students should ensure that they use storytelling and scripting elements such as emotion, pacing, and clarity in their dialogue.
• Students need to plan their recording locations (outside or in Sandbox Studios), considering potential sources of noise and how these will affect their takes.

Part 2: Recording Audio Clips: Clean, Noisy, and Distorted Takes

• Students use audio equipment (either loaned out equipment or Sandbox equipment) to record three versions of their dialogue: one clean, one with noticeable background noise, and one intentionally poor with pops and distortion.
• Students should demonstrate their understanding of microphone placement and techniques for minimizing or capturing noise, such as facing away from traffic or introducing interference.
• Students will need to ensure they save each of these recordings for part 3 of the assignment.

Part 3: Manual Audio Restoration and Editing

• Students import their audio files into editing software (either on their own devices or Sandbox workstations) and use manual tools such as EQ, noise reduction, and de-essing to improve each take.
• Students are encouraged to experiment with different filters and effects; this encourages students to expand their expertise in audio editing.
• Students export their best manual mixes for each take, preparing them for side-by-side comparison with AI results.

Part 4: AI Audio Enhancement and Comparative Analysis

• Students are now required to upload their raw original audio files to an AI plugin or Adobe Podcast - Enhance Speech, processing each take to generate AI-restored versions.
• Students are encouraged to use their AI literacy skills to try to improve the AI edited versions of the audio recordings through prompting and other options.
• Students will critically listen to and evaluate each version of the audio and identify AI capabilities and limitations in comparison to their manually edited versions.

Part 5: Reflection, Peer Feedback, and Assessment

• Students complete a short, written reflection comparing their manual and AI results, describing what surprised them and what they learned about each method.
• Students should identify in their reflections AI capabilities and limitations, such as loss of nuance or over-processing, and discuss ethical considerations in media restoration.

Summary

Partnered students write & record a 15-30 second piece of dialogue with Seneca’s audio equipment or Seneca Sandbox studios in 3 different types of audio quality (good, medium, & poor) and use their own post-production editing skills to see how their audio preservation compares to AI’s audio preservation.