---
title: "AI Filmmaking Workshop — Pre-Workshop Account Setup"
project: ai-filmmaking-workshop
type: checklist
created: 2026-04-29
version: 1.0
---

# AI Filmmaking Workshop — Pre-Workshop Setup Checklist

**Complete ALL items at least 2 days before the workshop.** Some services have email verification delays. Create accounts, verify emails, and confirm you can log in.

---

## ⚠️ Critical (Must Complete)

### 1. ChatGPT (or Claude)
- **URL:** https://chat.openai.com (or https://claude.ai)
- **Cost:** Free
- **Time:** 2 minutes
- **Steps:**
  1. Create account (Google/email)
  2. Verify email
  3. Start a new chat — type "Hello" to confirm it works
- **Workshop use:** Module 1 (script writing, ideation)

---

### 2. ElevenLabs (Voice Generation + Cloning)
- **URL:** https://elevenlabs.io
- **Cost:** Free (10,000 characters/month)
- **Time:** 5 minutes
- **Steps:**
  1. Create account → verify email
  2. Go to **VoiceLab** → **Add Voice** → **Instant Voice Cloning**
  3. Upload a ~1-minute clean voice sample (your own voice, or any clean recording)
  4. Name your voice → click **Add Voice**
  5. Go to **Speech Synthesis** → select your voice → type "Testing one two three" → click **Generate**
  6. Confirm audio plays correctly
- **Workshop use:** Module 3 (dialogue generation, voice cloning)
- **Tip:** Record your voice sample in a quiet room with your phone (voice memos app). Export as MP3/WAV.

---

### 3. Kling AI (Video Generation)
- **URL:** https://klingai.com
- **Cost:** Free (~66 credits/month, ~10-15 short generations)
- **Time:** 5 minutes + possible wait for approval
- **Steps:**
  1. Create account (may require phone number verification)
  2. Complete profile setup
  3. Go to **Text-to-Video** or **Image-to-Video**
  4. Try generating a short clip to confirm it works
- **Workshop use:** Module 2 (image-to-video generation from keyframes)
- **Backup:** Also create at https://app.runwayml.com (free trial: 125 credits one-time)

---

### 4. Sync.so (Lip Sync)
- **URL:** https://sync.so
- **Cost:** Free tier
- **Time:** 3 minutes
- **Steps:**
  1. Create account → verify email
  2. Go to **Dashboard** → **API Keys** → create a key (save it!)
  3. Go to **Studio** → try uploading a short video + audio to test
- **Workshop use:** Module 4 (lip sync — the core exercise)

---

### 5. CapCut (Video Editing — Desktop App)
- **URL:** https://www.capcut.com (download desktop version)
- **Cost:** Free (some features watermarked on export)
- **Time:** 5 minutes
- **Steps:**
  1. Download and install CapCut Desktop (Mac or Windows)
  2. Create account (TikTok/Google/email)
  3. Open app → create new project → import any video file to confirm it works
  4. Confirm you can see the timeline and basic editing tools
- **Workshop use:** Module 5 (video assembly, editing)

**Alternative (if you prefer pro-grade):** DaVinci Resolve — https://www.blackmagicdesign.com/products/davinciresolve (free, no watermarks, steeper learning curve)

---

## 📋 Recommended (Do If You Can)

### 6. Suno (Music Generation)
- **URL:** https://suno.ai
- **Cost:** Free (10 songs/day, 5 generations of 2 songs each)
- **Time:** 2 minutes
- **Steps:**
  1. Create account (Google/Discord/email)
  2. Type a prompt like "cinematic ambient background music, piano, strings, 60 seconds"
  3. Click **Create** → listen to the result
- **Workshop use:** Module 3 (background music generation)
- **Tip:** Disable public sharing if you want to keep your generations private.

---

### 7. HeyGen (AI Avatar / Talking Head)
- **URL:** https://heygen.com
- **Cost:** Free (1 minute/month)
- **Time:** 5 minutes
- **Steps:**
  1. Create account → verify email
  2. Go to **Create Video** → choose an avatar or upload your photo
  3. Type dialogue text → generate
  4. Preview the talking head result
- **Workshop use:** Module 4 (comparison — talking head vs cinematic lip sync)

---

### 8. Canva (Storyboarding / Design)
- **URL:** https://canva.com
- **Cost:** Free
- **Time:** 2 minutes
- **Steps:**
  1. Create account (Google/email)
  2. Search for "Storyboard template"
  3. Open any template to confirm it loads
- **Workshop use:** Module 1 (storyboard assembly)

---

### 9. Adobe Podcast Enhance (Audio Cleanup)
- **URL:** https://podcast.adobe.com/enhance
- **Cost:** Free (browser-based, no install)
- **Time:** 1 minute
- **Steps:**
  1. No account needed — just bookmark the URL
  2. Test with any voice recording: upload → wait for processing → download enhanced version
- **Workshop use:** Module 3 (dialogue cleanup)

---

## 🖥️ Demo Station Setup (Organizer Only)

If you're setting up a machine for MuseTalk and ComfyUI demos (requires GPU):

### MuseTalk Setup
```bash
# Prerequisites: Python 3.10+, CUDA 11.8+, 6GB+ VRAM GPU
git clone https://github.com/TMElyralab/MuseTalk.git
cd MuseTalk
pip install -r requirements.txt
# Download weights: https://huggingface.co/TMElyralab/MuseTalk
# Place in ./models/
python -m scripts.inference --input_video demo.mp4 --input_audio demo.wav
```

### ComfyUI Setup
```bash
git clone https://github.com/comfyanonymous/ComfyUI.git
cd ComfyUI
pip install -r requirements.txt
# Download a base model (e.g., SD 1.5) to models/checkpoints/
python main.py
# Open http://127.0.0.1:8188 in browser
```

### Figma Weave (Cloud, No Install)
- **URL:** https://www.figma.com/weave/
- Create a Figma account (free) → access Weave
- Test: create a simple node pipeline (text prompt → image generation)

---

## 📱 Day-of Checklist

Bring to the workshop:
- [ ] Laptop (charged + charger)
- [ ] Headphones/earbuds (for audio work)
- [ ] A 1-minute voice recording of yourself (for voice cloning)
- [ ] 2–3 high-quality photos of yourself or a "character" (for keyframe generation)
- [ ] A very short script idea (1–3 lines of dialogue)
- [ ] All accounts created and logged in (open each tab before arriving)
- [ ] Sync.so API key saved somewhere accessible
- [ ] CapCut or DaVinci Resolve installed and tested

---

## 🚨 Troubleshooting

| Problem | Solution |
|---------|----------|
| Kling AI won't let me sign up | Try Runway as backup: https://app.runwayml.com |
| ElevenLabs voice clone sounds bad | Re-record in a quieter room, closer to the mic. Use WAV not MP3. |
| Sync.so generation fails | Check video format (MP4, under 100MB). Check audio format (WAV, under 20MB). |
| CapCut watermark on export | Use DaVinci Resolve instead (completely free, no watermarks). |
| GPU not available for MuseTalk | Use Sync.so free tier instead — same quality, no GPU needed. |
| Any account requires phone verification | Use your phone. These services verify to prevent abuse. Your data isn't shared. |

---

## 📞 Questions Before the Workshop?

Contact: [Organizer Name] at [contact info]

---

*Setup time estimate: 30–45 minutes total. Do this at least 48 hours before the workshop to catch any issues.*