maketalk
A command-line tool to create professional video presentations with title cards and transcriptions.
Features
- Convert MOV files to MP4 with automatic dimension normalization
- Standalone EBU R128 loudness normalization for any video file
- Merge multi-part video sections
- Extract and transcribe audio using Yakety
- Generate beautiful title cards with customizable design
- Create final presentation video with title cards and content
Prerequisites
$3
- Node.js 14+
- FFmpeg
- jq
- Chrome/Chromium (for title card generation)
$3
- Yakety transcribe tool for automatic transcription (or set
YAKD_TRANSCRIBE_PATH environment variable)
- Without Yakety, maketalk will generate a template for manual title editing
Installation
Install globally:
``bash
npm install -g maketalk
`
Or run with npx:
`bash
npx maketalk
`
Usage
$3
Important: MOV files must follow a specific naming convention:
- Format: XX-name.mov where XX is a two-digit section number
- Examples:
- ✅ 01-introduction.mov
- ✅
- ✅
- ❌ (missing section number)
- ❌ todo-workflow.mov (missing section number)
$3
1. Place your MOV files in a directory with proper naming (see above)
2. Run the tool:
`bash
maketalk
`
3. Follow the prompts to generate transcriptions and create a prompt for title generation
4. Use claude-danger with the generated prompt to create title_cards.json
5. Continue the process:
`bash
maketalk --continue
`
$3
1. Place your MOV files in a directory with proper naming (see above)
2. Run the tool:
`bash
maketalk
`
3. The tool will generate:
- sections.json - Information about your video sections
- title_cards.json - Template with placeholder titles
4. Edit title_cards.json to add your own titles and descriptions
5. Continue the process:
`bash
maketalk --continue
`
6. (Optional) Level the audio of the final presentation:
`bash
maketalk --level-audio final_presentation.mp4
`
$3
If you need to re-run the process but already have converted videos:
`bash
This will keep your converted videos and re-do all subsequent steps
maketalk --resume-after-conversion
`
This is useful when:
- You want to regenerate transcriptions
- You need to create different title cards
- Something failed after the conversion step
$3
`
bash
maketalk --preview 01 "Claude Code" "The Future of Programming"
`
$3
`
bash
Level audio of any video file (in-place)
maketalk --level-audio video.mp4
`
$3
- --level-audio
: Apply EBU R128 loudness normalization to a single file (standalone operation)
- --continue: Continue from the claude-danger step after creating title_cards.json
- --resume-after-conversion: Resume processing after MOV to MP4 conversion (keeps converted videos)
- --preview <desc><code class="bg-muted px-1 py-0.5 rounded text-sm font-mono">: Preview a title card design<br />- </code>--help<code class="bg-muted px-1 py-0.5 rounded text-sm font-mono">: Show help information</p><p class="my-3"><h2 class="text-xl font-semibold mt-5 mb-3">Environment Variables</h2></p><p class="my-3">- </code>YAKD_TRANSCRIBE_PATH<code class="bg-muted px-1 py-0.5 rounded text-sm font-mono">: Path to the Yakety transcribe binary (default: </code>/Users/badlogic/workspaces/yakety/build/bin/transcribe<code class="bg-muted px-1 py-0.5 rounded text-sm font-mono">)</p><p class="my-3"><h2 class="text-xl font-semibold mt-5 mb-3">Output</h2></p><p class="my-3">The tool creates a </code>generated/<code class="bg-muted px-1 py-0.5 rounded text-sm font-mono"> directory with:<br />- </code>audio/<code class="bg-muted px-1 py-0.5 rounded text-sm font-mono">: Extracted audio files<br />- </code>transcriptions/<code class="bg-muted px-1 py-0.5 rounded text-sm font-mono">: Text transcriptions<br />- </code>title_cards/<code class="bg-muted px-1 py-0.5 rounded text-sm font-mono">: Generated title card images and videos<br />- </code>converted_videos/<code class="bg-muted px-1 py-0.5 rounded text-sm font-mono">: Processed MP4 files</p><p class="my-3">Final output: </code>final_presentation.mp4<code class="bg-muted px-1 py-0.5 rounded text-sm font-mono"></p><p class="my-3"><h2 class="text-xl font-semibold mt-5 mb-3">Audio Leveling</h2></p><p class="my-3">The </code>--level-audio <file><code class="bg-muted px-1 py-0.5 rounded text-sm font-mono"> option applies professional EBU R128 loudness normalization to any video file. This is a standalone operation that:</p><p class="my-3">- Normalizes to -16 LUFS (suitable for online video platforms)<br />- Updates the file in-place<br />- Preserves video quality (video stream is copied, not re-encoded)<br />- Shows before/after audio level analysis</p><p class="my-3">Example workflow:<br /></code>`<code class="bg-muted px-1 py-0.5 rounded text-sm font-mono">bash<br /><h1 class="text-2xl font-bold mt-6 mb-4">Create the presentation</h1><br />maketalk</p><p class="my-3"><h1 class="text-2xl font-bold mt-6 mb-4">Level the audio of the final output</h1><br />maketalk --level-audio final_presentation.mp4<br /></code>``</p><p class="my-3">Note: Audio leveling requires two passes - one to analyze and one to apply normalization.</p><p class="my-3"><h2 class="text-xl font-semibold mt-5 mb-3">License</h2></p><p class="my-3">ISC</p></div><div class="flex justify-center absolute inset-x-0 bottom-0 bg-gradient-to-t from-background via-background to-transparent pb-4 pt-16"><button data-slot="button" class="inline-flex items-center justify-center whitespace-nowrap text-sm font-medium transition-all disabled:pointer-events-none disabled:opacity-50 [&_svg]:pointer-events-none [&_svg:not([class*='size-'])]:size-4 shrink-0 [&_svg]:shrink-0 outline-none focus-visible:border-ring focus-visible:ring-ring/50 focus-visible:ring-[3px] aria-invalid:ring-destructive/20 dark:aria-invalid:ring-destructive/40 aria-invalid:border-destructive border bg-background shadow-xs hover:bg-accent hover:text-accent-foreground dark:bg-input/30 dark:border-input dark:hover:bg-input/50 h-8 rounded-md gap-1.5 px-3 has-[>svg]:px-2.5"><svg xmlns="http://www.w3.org/2000/svg" width="24" height="24" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="lucide lucide-chevron-down mr-1 h-4 w-4"><path d="m6 9 6 6 6-6"></path></svg>Show more</button></div></div></div><template id="P:3"></template><template id="P:4"></template></div></div></div><div hidden id="S:3"><div data-state="inactive" data-orientation="horizontal" role="tabpanel" aria-labelledby="radix-_R_aav5ubrb_-trigger-dependencies" hidden="" id="radix-_R_aav5ubrb_-content-dependencies" tabindex="0" data-slot="tabs-content" class="flex-1 outline-none mt-6"></div></div><script>$RS=function(a,b){a=document.getElementById(a);b=document.getElementById(b);for(a.parentNode.removeChild(a);a.firstChild;)b.parentNode.insertBefore(a.firstChild,b);b.parentNode.removeChild(b)};$RS("S:3","P:3")</script><div hidden id="S:4"><div data-state="inactive" data-orientation="horizontal" role="tabpanel" aria-labelledby="radix-_R_aav5ubrb_-trigger-versions" hidden="" id="radix-_R_aav5ubrb_-content-versions" tabindex="0" data-slot="tabs-content" class="flex-1 outline-none mt-6"></div></div><script>$RS("S:4","P:4")</script><script>$RC("B:1","S:1")</script></body></html>