DESCRIPT MASTERY

The Three-Panel Paradigm

Descript's interface is built around three core panels that work in harmony. Understanding how these panels interact is essential for efficient editing.

Left Panel - Script View: This is your primary editing interface. The transcript appears as editable text, where every word corresponds directly to audio/video. When you delete text here, you're removing that spoken content from your media. This is where 80% of your editing happens.

Center Panel - Canvas/Preview: Your visual workspace showing the current video frame or waveform. For video projects, this displays your composition with layers. For audio-only projects, this shows the waveform. The canvas updates in real-time as you edit the script.

Right Panel - Properties & Layers: Context-sensitive controls that change based on what you've selected. This panel houses audio effects, video effects, layer management, and composition properties. It's your detailed control center.

Script View vs Timeline View

Descript offers two editing modes, each optimized for different tasks. Knowing when to use each is crucial for maximum efficiency.

When to Use Script View (Primary Mode):

• Removing filler words, ums, ahs, and pauses
• Cutting entire sections of spoken content
• Reordering interview segments
• Finding and replacing words across the entire project
• Reading through content to identify weak sections
• Adding word-by-word emphasis or corrections

When to Use Timeline View:

• Precise audio fade adjustments at edit points
• Fine-tuning video transitions frame-by-frame
• Synchronizing B-roll with specific moments
• Adjusting music bed levels under dialogue
• Placing overlay graphics at exact timestamps
• Working with non-transcribed audio (sound effects, music)

Project Types and When to Use Each

Descript offers multiple project templates, each optimized for specific use cases.

Video Project: Use when your final output is video content. This enables the full canvas with composition layers, screen recording, webcam, and video file imports. Ideal for: YouTube videos, training content, social media videos, video podcasts.

Audio Project: Optimized for podcast workflows. Simpler interface focused on audio tracks and waveforms without video composition overhead. Ideal for: podcast episodes, audiobooks, audio-only interviews, radio segments.

Recording Project: For capturing new content directly in Descript. Choose this when starting a recording session rather than importing existing files. Supports screen recording, webcam, microphone, or system audio capture.

The Core Editing Paradigm

Understanding how text manipulation affects media is fundamental to Descript mastery. Every action in the script has a direct, predictable impact on your audio/video.

Delete Text = Remove Media: When you select words in the script and press Delete, Descript removes that spoken audio/video segment. The media before and after automatically connects. This is how you eliminate mistakes, false starts, and rambling sections.

Cut/Copy/Paste Text = Rearrange Media: Standard text editing commands work on media. Select a paragraph, cut it (Cmd/Ctrl + X), move your cursor to a new location, and paste (Cmd/Ctrl + V). You've just reordered your video content by manipulating text.

Find & Replace = Batch Editing: Use Cmd/Ctrl + F to find specific words across your entire transcript. When you replace them, Descript can regenerate audio using Overdub (covered in Module 3) or simply find all instances for manual review.

Selection Techniques for Efficiency

How you select text dramatically impacts editing speed. Master these selection patterns to edit faster.

Word Selection: Double-click any word to select it. This selects the word plus its associated timing. Useful for: removing single filler words ("um," "uh," "like"), replacing mispronunciations, isolating specific terms.

Sentence Selection: Triple-click to select an entire sentence. The selection respects natural speech boundaries. Useful for: removing complete thoughts, reordering key points, batch deletion of weak statements.

Range Selection: Click at start point, hold Shift, click at endpoint. This works across multiple paragraphs. Useful for: removing large segments, moving entire sections, selecting multi-speaker exchanges.

Multi-Selection (Advanced): Hold Cmd/Ctrl and click multiple non-contiguous words. This selects separate words that aren't adjacent. Useful for: removing multiple filler words at once ("um... and uh... you know"), isolating repeated words across a section.

Handling Silence and Pauses

Descript represents silence as gaps in the transcript. Learning to manage these gaps is essential for pacing control.

Viewing Silence: Enable "Show Gaps" in View menu (Cmd/Ctrl + Shift + G). Pauses appear as teal markers labeled with duration (e.g., "2.5s"). This reveals the pacing rhythm of your content.

Shortening Pauses: Click a gap marker to select it. Press Delete to remove entirely, or drag the marker edges to shorten the pause to your desired length. This tightens pacing without affecting the spoken content.

Adding Pauses: Place your cursor between words and press "Add Gap" or use the Add menu. Specify duration. This creates breathing room for dramatic effect or comprehension time.

Batch Gap Reduction: Select a large section of transcript, go to Edit → Shorten Word Gaps. Set maximum gap duration (try 1.2 seconds). Descript automatically reduces all pauses longer than this threshold, dramatically tightening pacing across entire segments.

Correcting Transcription Errors

Descript's AI transcription is highly accurate but occasionally misinterprets words. Corrections improve both readability and searchability.

Inline Correction: Click any word to edit it directly. The correction appears in the script but doesn't change the underlying audio. This is perfect for proper nouns, technical terms, or brand names that were transcribed phonetically.

When to Correct vs Leave: Correct if the word affects meaning or professional appearance in captions/transcripts. Don't waste time correcting minor transcription quirks that won't be visible to audiences (like "gonna" vs "going to").

Building Vocabulary: When correcting specialized terms that will recur in future projects, add them to Descript's vocabulary (right-click → Add to Vocabulary). This trains the AI to correctly transcribe these terms in future projects.

Speaker Labels and Organization

Multi-speaker content requires proper speaker labeling for clarity and advanced editing features.

Automatic Detection: Descript attempts to detect different speakers and labels them as Speaker 1, Speaker 2, etc. This works best with distinct voice characteristics and good recording technique (separate tracks per speaker).

Manual Assignment: Click any speaker label to rename it (e.g., "Sarah," "John"). Once renamed, click other sections labeled "Speaker 1" and reassign them to "Sarah." Descript learns voice patterns and improves future auto-detection.

Speaker-Based Editing: Once speakers are labeled, you can isolate content by speaker. Right-click speaker name → Select All [Speaker Name]. This selects everything that person said, enabling bulk operations like applying consistent audio processing or removing entire speaker tracks.

Using Comments for Collaboration

Comments enable async collaboration and self-review, crucial for client work and team projects.

Adding Comments: Select any text, click the comment icon (or Cmd/Ctrl + Shift + M). Type your note. Comments are timestamp-specific and visible to all project collaborators.

Comment Types: Use comments for different purposes: "Client approval needed here," "Find B-roll of this product," "Audio quality drops - re-record?," "Fact-check this claim," "Consider removing this section."

Resolution Workflow: Once addressed, mark comments as resolved. They disappear from active view but remain in comment history. This creates a clean audit trail of feedback and revisions.

Core Navigation Shortcuts

Memorizing these shortcuts transforms your editing speed from methodical to lightning-fast.

• Space: Play/Pause (works from anywhere in the interface)
• Cmd/Ctrl + Shift + V: Toggle between Script and Timeline view
• Cmd/Ctrl + /: Hide/show right panel for more script space
• Cmd/Ctrl + Shift + G: Show/hide word gaps (silence markers)
• Cmd/Ctrl + F: Find text
• Cmd/Ctrl + G: Find next instance
• Tab: Jump to next edit point or gap
• Shift + Tab: Jump to previous edit point or gap

Advanced Editing Shortcuts

These shortcuts handle complex editing operations in seconds.

• Cmd/Ctrl + Shift + R: Remove selected audio while keeping video (or vice versa)
• Cmd/Ctrl + D: Duplicate selected clip/layer
• Cmd/Ctrl + B: Split clip at playhead (timeline view)
• Cmd/Ctrl + Shift + D: Delete selected and close gap
• Cmd/Ctrl + Shift + M: Add comment at current position
• I: Set In point (timeline view)
• O: Set Out point (timeline view)
• Option/Alt + Delete: Ripple delete (removes selection and shifts everything after)

What is a Composition?

A composition in Descript is a visual container that holds and organizes all your video elements—video clips, images, text, shapes, and effects. Think of it as a Photoshop document for video: multiple layers stacked vertically, each layer representing a different visual element.

How Compositions Work: When you create a video project, Descript automatically creates a composition. The canvas (center panel) displays this composition. Every element you add—whether recorded video, imported footage, or graphics—becomes a layer within this composition.

Timeline Integration: Compositions exist at specific points in your timeline. You can have multiple compositions in a single project, each with different layouts. For example: intro composition (logo animation), main content composition (picture-in-picture), outro composition (call-to-action screen).

Layer Hierarchy: Layers stack from bottom to top. Bottom layers appear behind, top layers appear in front. This z-axis ordering determines visibility when elements overlap. Understanding this hierarchy is crucial for complex layouts.

Canvas Controls and Aspect Ratios

The canvas is your visual workspace where you arrange and preview all composition elements. Proper canvas configuration is essential for platform-specific content.

Accessing Canvas Settings: Click the composition name at the top of the canvas or go to Properties panel (right side). Here you control aspect ratio, resolution, and background properties.

Common Aspect Ratios and When to Use Each:

• 16:9 (1920x1080): YouTube, Vimeo, horizontal social posts, traditional video. This is the standard widescreen format.
• 9:16 (1080x1920): Instagram Reels, TikTok, YouTube Shorts, Stories. Vertical mobile-first format.
• 1:1 (1080x1080): Instagram feed posts, LinkedIn posts, Twitter videos. Square format maximizes mobile screen real estate.
• 4:5 (1080x1350): Instagram feed (vertical). Taller than square, more screen space than 1:1 without full 9:16.
• Custom: Any resolution for specific client requirements or unique platforms.

Background Management

Backgrounds set the foundation for your composition's visual identity. Professional content uses intentional backgrounds rather than default settings.

Solid Color Backgrounds: Click canvas, go to Properties → Background → Color. Choose from presets or use hex codes for brand colors. Ideal for: clean professional looks, corporate content, minimalist designs, when content should be the focus.

Gradient Backgrounds: Properties → Background → Gradient. Set start and end colors, adjust angle. Creates depth without distraction. Ideal for: modern aesthetics, tech content, when you need visual interest without competing with foreground elements.

Image Backgrounds: Drag image file onto canvas, then right-click → Send to Back (or drag to bottom layer). Adjust opacity if needed. Ideal for: branded content (company colors/patterns), topical content (relevant imagery), creating environmental context.

Video Backgrounds: Import video file, send to back layer, often applied with blur or opacity reduction. Ideal for: dynamic content, B-roll as environmental context, creating movement in static presentations.

The Layers Panel

The Layers panel (right side, below Properties) is your command center for complex compositions. Efficient layer management separates amateur from professional work.

Viewing Layers: If the Layers panel isn't visible, go to View → Show Layers (or Cmd/Ctrl + Shift + L). The panel lists all layers in your composition from top (foreground) to bottom (background).

Layer Operations:

• Renaming: Double-click any layer name to rename it. Use descriptive names: "Main_Camera," "ScreenRecording," "Logo_TopRight," not "Layer 1," "Layer 2."
• Reordering: Drag layers up/down in the panel to change stacking order. Top layers appear in front of bottom layers.
• Visibility Toggle: Click the eye icon to hide/show layers. Hidden layers remain in the project but don't appear in output.
• Locking: Click the lock icon to prevent accidental changes. Lock background layers or finalized elements.
• Grouping: Select multiple layers (Cmd/Ctrl + click), right-click → Group. Groups collapse for cleaner organization.

Layer Positioning and Sizing

Precise positioning creates professional layouts. Descript offers multiple methods for positioning layers, each optimized for different scenarios.

Manual Positioning (Canvas): Click and drag any layer on the canvas to reposition. Handles appear at corners and edges for resizing. Hold Shift while dragging to constrain proportions. This is intuitive but imprecise.

Numeric Positioning (Properties Panel): Select layer, go to Properties → Position & Size. Set exact X/Y coordinates and width/height in pixels. This provides pixel-perfect placement for professional consistency.

Alignment Tools: Select layer, go to Layout menu or right-click → Align. Options: Align Left, Center Horizontal, Align Right, Align Top, Center Vertical, Align Bottom. Use these to snap elements to canvas edges or center.

Preset Positions: Descript offers quick position presets in the Properties panel: Top Left, Top Center, Top Right, Middle Left, Center, Middle Right, Bottom Left, Bottom Center, Bottom Right. One-click positioning for common layouts.

Layer Effects and Properties

Layer-specific effects enhance visual quality and professionalism without requiring external editing tools.

Opacity: Select layer → Properties → Opacity slider (0-100%). Reduce opacity for subtle overlays, watermarks, or blended backgrounds. At 0%, layer is invisible (same as hiding). At 50%, layer is semi-transparent.

Corner Radius: Properties → Corner Radius. Adds rounded corners to any layer. Set to 10-20px for subtle softening, 999px for perfect circles. Ideal for: modern aesthetics, softening webcam edges, creating circular profile displays.

Border: Properties → Border. Add colored borders around layers. Set color, width (pixels), and opacity. Use for: emphasizing elements, creating separation from background, branded accent colors.

Shadow: Properties → Shadow. Add drop shadows for depth. Adjust X/Y offset (direction), blur (softness), and opacity. Subtle shadows (2-3px offset, 30% opacity) add professional depth without being obvious.

Rotation: Properties → Rotation. Rotate layer in degrees. Use for: creative compositions, fixing crooked source footage, dynamic transitions.

Template System Overview

Templates are pre-configured compositions that you can reuse across projects. They're the secret to consistent branding and rapid production. Build once, reuse infinitely.

What Templates Store: Templates save your entire composition setup—layer arrangement, positioning, sizing, effects, colors, fonts, and placeholder content. When you apply a template, it recreates this exact setup in your new project.

Template Use Cases: Brand consistency (same layout across all videos), rapid production (new video in minutes, not hours), client work (one template per client for their brand), platform-specific formats (YouTube template, Instagram template, TikTok template), recurring content (podcast intro/outro, weekly show format).

Creating Your First Template

Building effective templates requires strategic thinking about what varies versus what stays constant across videos.

Step-by-Step Template Creation:

1. Build Ideal Composition: Create a composition with your perfect layout. Add all layers: background, positioning boxes for video/images, text placeholders, logo, branding elements.
2. Use Placeholder Content: For varying content (main video, images), add temporary placeholders or leave empty layers with clear names like "INSERT_MAIN_VIDEO_HERE."
3. Lock Static Elements: Lock layers that should never change (background, logo position, brand colors). This prevents accidental modification.
4. Name Everything: Give every layer a descriptive name. Template users need to understand what goes where.
5. Test the Layout: Add sample content to each placeholder to ensure sizing/positioning works correctly.
6. Save as Template: Go to File → Save as Template. Give it a descriptive name and category.

Applying and Managing Templates

Knowing how to efficiently apply and modify templates accelerates your workflow dramatically.

Applying Templates: Create new project → Click "Start from template" → Select your template → Click "Use Template." Descript creates a new project with all template layers and settings intact.

Updating Templates: Open a project using the template → Make improvements → File → Update Template. This updates the master template, but doesn't affect existing projects (they remain unchanged).

Template Libraries: Organize templates into categories: "YouTube Formats," "Instagram Content," "Client Work," "Podcast Layouts." This organization prevents template overwhelm as your library grows.

Sharing Templates: Export templates (File → Export Template) to share with team members or clients. They can import (File → Import Template) to maintain consistency across distributed teams.

Video Layer Management

Video layers are your primary content carriers. Understanding how to manipulate them efficiently is essential for complex productions.

Adding Video Layers: Drag video files from your computer directly onto the canvas. Alternatively, record directly in Descript (screen + webcam) and layers auto-generate. Each video becomes a separate layer you can position independently.

Trimming Video Layers: In Timeline view, each video layer has independent timing. Drag the edges to trim start/end points. This is crucial for picture-in-picture scenarios where webcam and screen recording need different in/out points.

Fill vs Fit: Properties → Scaling. "Fill" crops video to fill the entire layer space (may crop edges). "Fit" shows the entire video (may show letterboxing). Choose based on content importance.

Crop Tool: Select layer → Properties → Crop. Manually adjust which portions of the video are visible. Use for: removing unwanted screen edges, focusing on specific areas, creating custom aspect ratios from source footage.

Image Layers and Graphics

Static images serve as logos, backgrounds, callouts, product shots, and brand elements. Strategic image layer use elevates production value.

Supported Formats: PNG (with transparency - ideal for logos/graphics), JPEG (no transparency - backgrounds/photos), GIF (animated - use sparingly), SVG (vector - perfect for logos, scales without quality loss).

Transparency Handling: PNG files with transparent backgrounds automatically blend with your composition. This is crucial for logos, brand elements, and callout graphics. Position them over video without white boxes.

Layer Duration: Image layers can span your entire video or appear briefly. In Timeline view, drag image layer edges to control when it appears. Use for: logo watermarks (full duration), product callouts (brief appearance), slide deck images (specific segments).

Text Layers and Typography

Text layers communicate information, emphasize points, and establish brand identity. Professional text layer usage separates amateur from polished content.

Adding Text: Click "T" in toolbar or Insert → Text. A text box appears on canvas. Type your content, then configure styling in Properties panel.

Typography Controls:

• Font: Choose from system fonts. For brand consistency, use the same 2-3 fonts across all content.
• Size: Measured in pixels. Minimum 36px for mobile visibility, 48-72px for emphasis.
• Weight: Light, Regular, Bold, etc. Use bold for emphasis, regular for body text.
• Color: Click color picker for exact brand colors (use hex codes for consistency).
• Alignment: Left, center, right. Center for titles/emphasis, left for body text.
• Line Height: Space between text lines. 1.2-1.4x for tight layouts, 1.5-2x for readability.

Shape Layers for Design

Shapes create visual structure, backgrounds for text, highlighting areas, and geometric design elements.

Available Shapes: Rectangle, Circle, Line. Access via Insert menu or toolbar. Each shape is a vector layer that scales without quality loss.

Rectangle Uses: Backgrounds for text (lower thirds, title cards), full-screen overlays with transparency (darkening video for text readability), borders and frames, geometric design elements.

Circle Uses: Highlighting specific areas (red circle around UI element), profile frames (circular mask for headshots), bullet points and decorative elements, loading animations.

Line Uses: Underlines for emphasis, separators between sections, arrows and pointers, creating geometric compositions.

What is Overdub?

Overdub is AI-powered voice synthesis that creates realistic speech from text input. You train the AI on your voice (or a selected voice), then generate new audio by typing. The AI matches tone, cadence, and accent, producing natural-sounding speech indistinguishable from real recordings in most contexts.

Core Use Cases: Fixing mistakes without re-recording (changed a word wrong, mispronounced something), updating content with new information (product names, statistics, dates), removing filler words cleanly (replace "um" with smooth audio), script corrections discovered in post-production, creating consistent narration across multiple videos, multilingual content with accent consistency.

How it Works Technically: The AI analyzes your voice training data to learn your speech patterns, pronunciation tendencies, pitch range, speaking rhythm, and accent characteristics. When you type new text, it generates audio matching these learned patterns. The result sounds like you recorded those exact words naturally.

Creating Your Voice Clone

The voice training process captures your unique vocal characteristics. Quality training data determines output realism.

Training Requirements: Descript requires 10 minutes of clean, clear speech for basic training. For best results, record 30+ minutes across multiple sessions. The AI needs variety: different sentences, emotions, speaking speeds, and contexts to capture your full vocal range.

Recording Environment Setup:

• Quiet Space: No background noise (AC, traffic, computer fans). Background sounds get baked into your voice model.
• Consistent Microphone: Use the same mic you'll use for content. Voice characteristics change between microphones.
• Proper Distance: 6-8 inches from mic. Consistent positioning across all training recordings.
• Natural Delivery: Speak normally, not in "announcer voice." The AI clones your natural speaking pattern.
• Emotional Range: Include enthusiastic sections, calm explanations, and conversational tones. Varied training = versatile output.

Training Process Step-by-Step

Navigate to Account Settings → Overdub → Create New Voice. Follow the guided process.

1. Name Your Voice: Use descriptive names like "John_Professional" or "Sarah_Casual" if creating multiple versions.
2. Record or Upload: Either record directly in Descript or upload pre-recorded training files. Upload is better for controlled studio recordings.
3. Review Recording Quality: Descript analyzes audio quality. If training data has issues (noise, inconsistent levels), you'll receive warnings. Address these before proceeding.
4. Processing Time: AI training takes 5-30 minutes depending on audio length. You'll receive email notification when complete.
5. Voice Preview: Test your new voice with sample text before using in projects. This ensures quality meets expectations.

Stock AI Voices

Descript offers pre-trained professional voices when you don't want to use your own voice or need variety.

When to Use Stock Voices: Character voices for storytelling, professional narration without personal branding, placeholder audio during editing (replace with real voice later), multilingual content, client work where you're behind the scenes, anonymous content creation.

Available Voice Types: Male and female voices, multiple accents (American, British, Australian), various ages and tones (authoritative, friendly, enthusiastic), professional narrator styles, conversational styles.

Stock Voice Limitations: Cannot be customized or retrained, limited emotional range compared to custom voices, same voice used by other Descript users, less authentic for personal brand content.

Word-Level Replacements

The most common Overdub use case—replacing individual words or phrases without disturbing surrounding audio.

Process: In your transcript, select the word(s) to replace. Type the correction directly. Click the Overdub button (or right-click → Overdub Selection). Choose your voice. Generate. The AI creates that word in your voice, matching the surrounding audio's tone and pace.

Best Practices for Natural Replacements:

• Include surrounding words for context: Replace "the blue button" not just "blue" for better flow
• Match original sentence structure: Keep similar word count and rhythm
• Preview before finalizing: Listen to the edit in context
• Adjust surrounding gaps: Overdub may need slight gap adjustment before/after for natural timing

Sentence and Paragraph Generation

Beyond word replacement, Overdub generates entirely new sentences—perfect for adding content discovered missing during editing.

When to Generate New Content: Forgot to mention key point during recording, need transitional sentences between sections, client requests additional explanation, creating updated version of old content, script expansion after initial recording.

Writing for Overdub: Write the way you naturally speak—use contractions ("don't" not "do not"), include natural pauses with punctuation, avoid complex compound sentences, match your typical speaking vocabulary, consider emphasis (Overdub respects sentence structure for natural flow).

Overdub Quality Optimization

Achieving natural-sounding results requires understanding Overdub's strengths and limitations.

What Works Best: Short phrases (under 30 words per generation), natural sentence structure, your typical vocabulary, moderate pacing, standard pronunciation.

What's Challenging: Long complex sentences (over 50 words), extreme emotional range (yelling, whispering), rapid speech or unusual pacing, made-up words or unusual names, heavy regional dialect variations.

Improving Output Quality:

• Break Long Sentences: Generate in shorter chunks, then combine
• Add Pronunciation Guides: Use phonetic spelling in your typed text for difficult words
• Regenerate with Variations: Try different word orders or phrasing if first attempt sounds unnatural
• Blend with Real Audio: Use Overdub for small corrections within real recordings rather than full synthetic narration

Ethical Overdub Use

AI voice cloning carries ethical responsibilities. Professional use requires transparent practices.

Your Own Voice Ethics: Using Overdub on your own voice for your own content is ethically clear. You're simply improving efficiency. Disclosure isn't legally required but builds audience trust when appropriate.

Client Work Ethics: When creating content for clients using your voice, inform them you use AI assistance for corrections. Include this in your service agreement. Most clients appreciate the efficiency; transparency prevents future issues.

Third-Party Voices: Never create Overdub voices of other people without explicit written permission. This includes celebrities, colleagues, or clients. Legal and ethical violations can result in serious consequences.

Disclosure Best Practices: For content representing yourself, subtle disclosure maintains authenticity: "Edited using AI-assisted tools," or acknowledgment in video description. For content where voice is critical to message (testimonials, expert opinions), err toward full transparency.

The Script-First Workflow

Advanced creators write scripts, generate complete Overdub narration, then add visuals. This inverts traditional video production.

Traditional Workflow: Write outline → Record video → Edit → Realize script issues → Re-record → Final edit. Time: 8-12 hours for 10-minute video.

Script-First Workflow: Write complete script → Generate Overdub narration → Review/refine script → Add screen recording/visuals → Final polish. Time: 4-6 hours for 10-minute video.

Advantages: Perfect script before any visual work, no re-recording for script improvements, consistent pacing and tone, ability to A/B test different narration approaches, faster iteration during client approval process.

Hybrid Recording Approach

Combine real recordings with strategic Overdub usage for optimal quality and efficiency.

When to Use Real Voice: Emotional content (personal stories, motivational speeches), spontaneous reactions and authentic moments, live interviews and conversations, content where "realness" is the value proposition.

When to Use Overdub: Factual corrections (dates, names, statistics), script refinements discovered in editing, filler word replacements, retakes of specific sentences (rather than full sections), updates to existing content.

Hybrid Workflow: Record primary content naturally and authentically. During editing, use Overdub for specific corrections and improvements rather than re-recording entire sections. This preserves authentic energy while eliminating inefficient retakes.

Content Update Strategy

Overdub enables sustainable evergreen content that stays current without complete re-production.

The Evergreen Problem: Tutorial videos become outdated (UI changes, new features, updated processes). Traditional solution: Complete re-recording. Cost: High. Frequency: Limited.

Overdub Solution: Update only changed sections. Replace outdated sentences with current information. Maintain 90% of original content while keeping video current.

Update Workflow: Review content quarterly, identify outdated sections, rewrite affected sentences with current information, generate new audio with Overdub, replace old segments, re-export video.

What Studio Sound Does

Studio Sound is AI-powered audio enhancement that analyzes your recording and automatically removes unwanted elements while preserving voice quality. It handles multiple issues simultaneously: background noise (HVAC, traffic, computer fans), room echo and reverb, inconsistent levels, mouth clicks and breathing sounds, low-quality microphone characteristics.

How It Works: The AI was trained on thousands of professional studio recordings paired with their raw source files. It learned to identify the difference between "amateur" and "professional" audio, then applies transformations to make your recording match studio characteristics. This happens in real-time processing with one click.

When Studio Sound Excels: Home office recordings with AC noise, interviews recorded in non-studio environments, podcast recordings with varying microphone quality, screen recordings with computer fan noise, outdoor recordings with wind or ambient sound, content recorded on budget equipment.

Studio Sound Limitations: Cannot fix severely distorted audio (clipping), cannot separate overlapping speakers effectively, cannot recover completely inaudible content, works best on speech—music/sound effects may sound unnatural, extreme processing can introduce slight artifacts.

Applying Studio Sound

Studio Sound can be applied at the track level or to individual clips. Understanding when to use each approach optimizes results.

Track-Level Application: Select the entire audio track in timeline, click the Studio Sound button in properties panel. This processes all audio in that track consistently. Ideal for: single-speaker recordings, consistent recording environment, podcast monologues, tutorial voiceovers.

Clip-Level Application: Select specific audio clips that need enhancement while leaving others untouched. Right-click clip → Apply Studio Sound. Ideal for: multi-speaker podcasts (different recording quality per person), recordings with mixed environments, interviews where only guest audio needs fixing, selectively enhancing problematic sections.

Intensity Control: Studio Sound offers intensity slider (0-100%). Default is 100% (maximum processing). Reduce to 50-75% if full processing sounds over-processed or unnatural. Higher intensity = more aggressive noise removal but potential for artifacts.

Studio Sound Best Practices

Maximizing Studio Sound effectiveness requires strategic application and awareness of its processing characteristics.

• Apply Early in Workflow: Run Studio Sound before other audio processing (EQ, compression). This provides the cleanest foundation for further refinement.
• Monitor for Artifacts: Listen carefully at normal playback speed. If voice sounds "underwater" or has digital artifacts, reduce intensity or try clip-level application.
• Compare Before/After: Toggle Studio Sound on/off multiple times. Sometimes subtle improvements are hard to notice until you hear the original again.
• Consider Content Type: Aggressive enhancement suits educational content where clarity matters most. Lighter touch preserves authenticity for personal/storytelling content.
• Save Processing Time: Studio Sound processes audio, which takes time for long recordings. Apply to clips rather than full tracks if only portions need enhancement.

Volume and Level Management

Consistent audio levels are essential for professional content. Descript offers multiple tools for volume control.

Clip Volume: Select clip, adjust volume slider in properties panel. Measured in dB (decibels). 0dB is original level, negative values reduce, positive values boost. Use for: balancing multiple speakers, adjusting music bed levels, fixing quiet sections.

Track Volume: Adjust entire track's level uniformly. Affects all clips in that track. Use for: overall mix balance, creating consistent output levels, managing background music tracks.

Normalization: Select audio → Right-click → Normalize. This analyzes peak levels and adjusts to target loudness. Ideal for: balancing interview participants, evening out level variations, preparing audio for specific platforms (podcasts have different standards than YouTube).

EQ (Equalization) Fundamentals

EQ shapes the tonal balance of audio—boosting or cutting specific frequency ranges. Strategic EQ makes voices clearer and more pleasant.

When to Use EQ: After Studio Sound processing, when voice lacks clarity or warmth, to reduce harsh frequencies, to differentiate multiple speakers (each gets slightly different EQ), to compensate for microphone characteristics.

Frequency Ranges Explained:

• Low (80-250 Hz): Warmth and body. Boost for thin voices, cut for muddy/boomy sound.
• Low-Mid (250-500 Hz): Fullness. Excessive can sound boxy. Cut slightly if voice sounds muffled.
• Mid (500-2000 Hz): Voice clarity and presence. This is where speech intelligibility lives.
• High-Mid (2000-4000 Hz): Articulation and definition. Boost for clarity, cut if harsh.
• High (4000-8000 Hz): Brightness and air. Boost for crisp sound, cut for softness.

Compression for Consistency

Compression evens out volume variations—loud parts become quieter, quiet parts become more audible. This creates consistent, professional-sounding audio.

Why Compression Matters: Natural speech varies dramatically in volume. Emphasize words spike louder, mumbled phrases drop quiet. Compression narrows this range, making everything audible without constant volume riding.

When to Compress: After Studio Sound and EQ, for content with dynamic speaking (enthusiasm creates volume spikes), to match levels between multiple speakers, for podcast/video where consistent intelligibility matters, before final limiting.

Compression Settings: Select track → Properties → Compression. Ratio controls aggressiveness (3:1 for gentle, 6:1 for obvious, 10:1 for limiting). Threshold determines when compression engages (-20dB for constant compression, -10dB for peaks only). Attack/Release control how fast compression responds (medium settings work for most voice).

Limiting and Final Polish

Limiting is the final safety net—preventing distortion while maximizing loudness. It's the last process before export.

What Limiting Does: Acts as aggressive compression that absolutely prevents audio from exceeding a set ceiling. Think of it as a brick wall—nothing gets through. Use for: preventing clipping (distortion), achieving platform loudness standards, catching unexpected peaks, final quality control.

Limiter Settings: Ceiling: -1dB to -0.5dB (prevents digital clipping with small safety margin). This is the absolute maximum level. Everything above this ceiling gets compressed down. Apply to master output, not individual tracks.

Music and Background Audio

Background music enhances mood but must be balanced carefully to avoid competing with speech.

Ducking: Automatically lowers music volume when speech is present, raises it during silence. This maintains musical presence without interference. Enable: Select music track → Properties → Ducking → Choose voice track to follow.

Manual Music Mixing: Set base music level at -20dB to -18dB below voice. During intro/outro without speech, boost music to -6dB to -3dB for impact. Use fade in/out for smooth transitions (2-3 seconds typical).

Music Selection Guidelines: Instrumental only (vocals compete with speech), consistent energy level (avoid dynamic builds during speech), appropriate mood (match content tone), royalty-free or licensed (avoid copyright issues).

Multi-Speaker Balance

Interviews and conversations require careful level matching between speakers for professional results.

The Reference Speaker Method: Choose the speaker with best audio quality as reference at 0dB. Match all other speakers to this reference by ear. Slight differences are acceptable (humans don't speak at identical volumes), but perception should be balanced.

Compensation Strategies: If one speaker is significantly quieter after level matching, apply additional compression to that track only. This brings up quiet parts while controlling peaks. If one speaker sounds thin compared to others, use EQ to add low-end warmth matching the fuller speaker.

Fixing Problem Audio

Sometimes audio has specific issues that require targeted solutions beyond Studio Sound.

Room Echo/Reverb: Studio Sound handles moderate reverb. For extreme cases: use additional de-reverb plugin, or re-record in better environment if critical content.

Plosives (P-Pops): Harsh "p" and "b" sounds. Solution: Use de-esser targeting low frequencies (80-150 Hz), or manually reduce volume of specific plosive instances, or high-pass filter more aggressively (100 Hz instead of 80 Hz).

Sibilance (Harsh S Sounds): Overly sharp "s" sounds. Solution: De-esser targeting 6000-8000 Hz, or EQ cut at sibilant frequencies.

Clipping/Distortion: Audio recorded too loud, causing digital distortion. Studio Sound cannot fix true clipping. Prevention is critical—monitor levels during recording, keep peaks below -6dB. Distorted audio requires re-recording.

Screen Recording Setup

Descript's screen recorder captures screen, webcam, and microphone simultaneously—ideal for tutorials, demos, and presentations. Unlike standalone screen recorders, recordings immediately become editable projects.

Starting a Recording: File → New Recording (or Cmd/Ctrl + Shift + R). Choose recording sources: screen only, screen + webcam, webcam only, or audio only. Select microphone and audio sources. Configure screen area (full screen, window, or custom area).

Recording Source Options:

• Full Screen: Captures entire display. Use for: software demos, presentations, anything requiring full context.
• Window: Captures specific application window. Use for: focused tutorials, when other windows contain sensitive info.
• Custom Area: Define specific screen region. Use for: hiding taskbars/menus, focusing on specific UI areas.
• Webcam: Simultaneously records your face. Position and size adjustable post-recording.

Recording Best Practices

Quality recordings start with proper preparation and technique, not post-production fixes.

Pre-Recording Checklist:

• Close Unnecessary Apps: Prevent notification popups and reduce CPU load for smooth recording.
• Hide Personal Info: Close email, Slack, anything with private information visible.
• Prepare Screen: Open applications you'll demonstrate, arrange windows optimally.
• Clear Desktop: Cluttered desktops appear unprofessional. Use clean background or folder organization.
• Zoom Appropriately: Text should be readable. Increase browser/app zoom to 125-150% if needed.
• Disable Notifications: Turn on Do Not Disturb mode to prevent interruptions.

Screen Recording + Webcam Layout

When recording screen and webcam simultaneously, Descript creates separate layers you can position independently.

Default Behavior: Screen recording becomes full-frame layer. Webcam appears as picture-in-picture in corner. Both layers are fully editable—reposition, resize, or hide as needed.

Common Layout Patterns:

• Corner Webcam: Screen full-frame, webcam small in corner (bottom-right typical). Use for: software tutorials where screen is primary focus.
• Side-by-Side: Screen takes 70% width, webcam 30% on side. Use for: presentations, interviews, when presenter presence is important.
• Webcam-Focused: Webcam full-frame, screen as small reference. Use for: reaction videos, commentary content.
• Dynamic Switching: Toggle between layouts during video. Use for: varied content needs within single recording.

System Audio Capture

Recording system audio captures sounds playing on your computer—essential for demos involving audio playback, video reviews, or music production tutorials.

When to Enable System Audio: Demonstrating audio features (music apps, sound design), reviewing videos/podcasts (capture their audio), recording webinars/presentations (capture presenter audio), game recordings (capture game sound).

When to Disable System Audio: General tutorials (unnecessary), when you want only your narration (cleaner audio), recording in noisy environment (system audio picks up notification sounds).

System Audio as Separate Track: Descript records system audio on dedicated track, independent from microphone. This enables individual volume control and processing—critical for balancing narration over demonstration audio.

Sharing and Permissions

Descript's collaboration features enable real-time teamwork without file exports, version confusion, or email chains.

Sharing Methods: Click Share button (top-right) to access sharing options. Can Share (edit permissions), View Only (review without editing), or Publish (public link). Each method serves different collaboration needs.

Permission Levels Explained:

• Owner: Full control. Can edit, delete, share, manage permissions. Usually the project creator.
• Can Edit: Full editing access. Can modify content, add comments, export. Cannot delete project or change permissions.
• Can Comment: Read-only with commenting. Can view, play, add comments. Cannot edit content. Ideal for clients/stakeholders.
• Can View: Read-only without commenting. Can view and play only. Useful for final review before publishing.

Comments for Feedback

Comments enable precise, timestamped feedback without back-and-forth emails or vague notes.

Adding Comments: Select text in transcript or click specific timestamp in timeline. Click comment icon or press Cmd/Ctrl + Shift + M. Type your comment. Comment appears at exact location for context.

Comment Types and Uses:

• Edit Notes: "Remove this section" or "Rewrite for clarity"
• Questions: "Is this data current?" or "Can we include B-roll here?"
• Approvals: "This section looks great" or "Approved for final"
• Technical Issues: "Audio quality drops here" or "Screen too small to read"
• Content Requests: "Add transition here" or "Include link in description"

Version History and Recovery

Descript automatically tracks all project changes. Access previous versions at any time without manual saving or file duplication.

Accessing Versions: File → Version History. See timeline of all changes with timestamps and author attribution. Preview any version. Restore previous version if needed.

When Version History Saves You: Client changes mind after you've made extensive edits (revert to earlier state), accidental deletion of content (recover from before deletion), testing different edits (revert if new approach doesn't work), tracking who made specific changes (team accountability).

Publishing and Public Sharing

Publish creates shareable link for public viewing—ideal for client presentations, portfolio pieces, or getting feedback from non-Descript users.

Publishing Options: Click Publish button. Choose what viewers can see: video only, video + transcript, allow downloading, enable comments. Generate unique link. Update published version anytime without changing link.

Published Link Use Cases: Client final review (send link instead of file), portfolio samples (showcase work publicly), stakeholder presentations (no login required), social proof (embed on website), team updates (share progress without granting edit access).

Record-to-Publish Workflow

Descript's integrated environment enables record-to-publish workflows that are impossible with fragmented tool chains.

Traditional Fragmented Workflow: Record (OBS/ScreenFlow) → Transfer files → Import to editor → Sync audio → Edit → Export → Send for review → Email feedback → Re-import → Make changes → Re-export → Final delivery. Time: 8-12 hours for 10-minute video.

Descript Integrated Workflow: Record → Auto-transcribe → Edit in same interface → Share link for review → Apply feedback in same project → Export final. Time: 3-5 hours for 10-minute video.

Multi-Platform Content Strategy

Create once in Descript, optimize for multiple platforms without separate projects or tool chains.

Single-Source Strategy: Record/edit master version in Descript. Duplicate project for each platform. Adjust canvas size and layer positioning per platform. Export all versions from same source material.

Understanding Export Settings

Export settings determine final file quality, size, and compatibility. Wrong settings waste upload time, reduce quality, or cause playback issues.

Key Export Parameters:

• Resolution: Pixel dimensions (1920x1080, 3840x2160, etc.). Higher = better quality but larger files.
• Frame Rate: Frames per second (24, 30, 60 fps). Match source footage or platform requirements.
• Codec: Compression method (H.264, H.265). H.264 = universal compatibility, H.265 = better compression.
• Bitrate: Data rate determining quality/size balance. Higher = better quality, larger file.
• Audio Settings: Codec (AAC standard), bitrate (128-320 kbps), sample rate (48kHz standard).

Descript Export Process

Descript simplifies exports with platform presets and custom control for advanced users.

Basic Export: File → Export → Choose format (Video or Audio). Select preset (YouTube, Instagram, etc.) or Custom. Choose destination and filename. Click Export. Descript processes and saves file.

Platform Presets: Descript includes optimized presets for major platforms. These handle resolution, codec, and bitrate automatically. Use presets for: standard uploads, when unsure about technical settings, fast exports without configuration.

Custom Settings: Advanced users can override presets. Access: Export → Custom Settings. Adjust resolution, frame rate, codec, bitrate manually. Use for: specific client requirements, non-standard platforms, file size optimization, maximum quality for archival.

Caption and Subtitle Export

Captions improve accessibility and engagement. Descript generates captions automatically from transcripts.

Caption Export Options: Burned-in (visible in video, cannot be disabled), SRT file (separate file, user-controllable), VTT file (web video standard), embedded (included in video file, toggle-able on platforms supporting captions).

When to Use Each: Burned-in captions for social media (auto-play without sound, captions ensure engagement). SRT/VTT files for YouTube, Vimeo (professional, accessible, multiple languages possible). Embedded captions for accessibility compliance, professional video platforms.

Quality Control Before Export

Final review prevents embarrassing mistakes and ensures professional delivery.

Pre-Export Checklist:

• Full Playthrough: Watch entire video start to finish. Check for awkward cuts, audio glitches, visual errors.
• Audio Levels: Consistent volume throughout. No clipping (red meters). Background music balanced under voice.
• Transcript Accuracy: All text spelled correctly (affects captions/SEO). Proper punctuation and capitalization.
• Visual Elements: Graphics appear at correct times. Layers positioned properly. No unintended elements visible.
• Branding: Logo present if required. Colors match brand guidelines. Consistent styling throughout.
• Export Settings: Correct resolution for platform. Appropriate aspect ratio. Captions configured correctly.

Service Packaging Strategy

Successful Descript businesses offer clear, value-based packages rather than hourly rates. Packages set expectations and improve profitability.

The Three-Tier Model: Offer three service levels (Basic, Professional, Premium) with clear value differentiation. This gives clients choice while steering most toward profitable middle tier.

Retainer Model for Recurring Revenue

Retainers provide predictable income and deeper client relationships. Monthly packages create stable business foundation.

Retainer Benefits: Guaranteed monthly income, priority scheduling, deeper understanding of client needs, reduced sales effort (maintain vs acquire), compound efficiency (templates reused monthly), client loyalty and long-term relationships.

Scaling Through Systems

Move from trading time for money to building scalable systems that increase income without proportional time increase.

Scalability Strategies: Template-based workflows (reduce per-project time), documented processes (enable delegation), tool mastery (work faster), selective client acceptance (higher-value projects only), automated communication (reduce admin time), batch production (efficiency gains).

Client Acquisition Strategies

Consistent client flow requires systematic lead generation and conversion processes.

Outbound Strategies: Direct outreach to potential clients, portfolio showcasing, cold email campaigns (personalized), LinkedIn connection + value-first engagement, partnership with complementary services (web designers, marketers).

Inbound Strategies: YouTube tutorials demonstrating expertise, social media presence (show before/afters), content marketing (blog about video production), SEO optimization (rank for "video editor near me"), referral program (incentivize existing clients).

Congratulations on Completing Descript Mastery

You've mastered the complete Descript ecosystem—from text-based editing fundamentals to advanced composition, AI voice technology, audio enhancement, collaboration workflows, and monetization strategies. You now possess skills that separate professional content creators from amateurs.

What You've Learned:

• Module 1: Text-based editing foundation that makes you 10x faster than traditional editors
• Module 2: Advanced composition and layer management for broadcast-quality visuals
• Module 3: AI voice cloning and Overdub for eliminating re-recording waste
• Module 4: Studio Sound and audio processing for professional-quality output
• Module 5: Screen recording and collaboration for seamless client workflows
• Module 6: Export optimization and business systems for sustainable income

Final Thoughts: From Skills to Income

Technical Descript mastery is only half the equation. Your success depends on positioning these skills as valuable services that solve real business problems.

Remember: Clients don't buy video editing—they buy: faster time-to-market for their content, professional brand image, engagement with their audience, freedom from production bottlenecks, scalable content systems.

Your Competitive Advantages: Speed (text-based editing is 5-10x faster), collaboration (real-time feedback vs email chains), versatility (recording + editing + audio in one tool), adaptability (quick updates via Overdub), professionalism (broadcast-quality output from home studio).

The Descript ecosystem eliminates traditional video production friction. You've learned to leverage this advantage. Now execute consistently, deliver exceptional results, and build systems that scale. Your first $10,000 month is not just possible—it's probable with proper execution.

Thank you for completing Descript Mastery. Your journey from learner to professional starts now. Go create exceptional content, deliver outstanding value, and build the business you deserve.