Commit 4b2b7b

2025-11-03 09:42:27 Melisha Dsouza: Updated
Projects/Flowcast.md ..
@@ 183,10 183,10 @@
e. Video output along with metadata is saved
- f. AI Agents
+ f. **AI Agents**
i. Analysis for the video by AI
- ii. Generate data for events:
+ ii. **Generate data for events:**
1. Timestamps
2. Locations
3. Type
@@ 206,18 206,14 @@
## **8\. AI Parts of the Challenge**
1. Code understanding and generating vector embedding and metadata for the flows
-
2. Video analysis (events, transitions, etc)
-
3. Video metadata generation
-
4. Transcripts, subtitle, voiceover generation
-
5. FFMPEG script/command generation
## **9\. Team plans and goals**
- 1. First 2 weeks goals to achieve:
+ 1. **First 2 weeks goals to achieve: **
a. Finalise the backend architecture (monolith/microservices, vector db, cache, docs, ai tools)
@@ 227,7 223,7 @@
d. Have some testing scenarios for the qa ready
- 2. First month goals to achieve:
+ 2. **First month goals to achieve:**
a. Have all the public and private apis ready
@@ 247,24 243,16 @@
## **11\. Technology Stack**
-
1. **Frontend:** React
-
2. **Backend:** Microservice of Java and Python
-
3. **LLM:** Gemini 2.5-FLash, gemini-embedding-001
-
4. **AI Orchestration:** LangChain
-
5. **Integrations:** FFMPEG, Automations
-
6. **DB:** Postgres (with pgvector)
-
8. **Deployment:** Docker, Gitlab CI
## **12\. Multi-Model Video Analysis Limits: The 6-Minute Test Case**
-
| Feature / Limit | Gemini API / Vertex AI (Developer/Enterprise) | Gemini Apps (Consumer Chat) - Free Tier | Gemini Apps (Consumer Chat) - AI Pro/Ultra | Workaround for Long Videos (> 1 Hour) |
| -------------------------- | --------------------------------------------- | --------------------------------------------------- | ------------------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------- |
| Max Single Video Duration | ≈ 45 mins (with audio) ≈ 1 hour (no audio) | 5 minutes (Total length across all uploaded videos) | 1 hour (Total length across all uploaded videos) | Segmentation & Chaining: Split into 45-minute chunks, analyze, and feed the summary of the previous chunk to the next prompt. |
@@ 290,3 278,46 @@
| **Gemini API / Vertex AI** | ≈ 1 hour (Default resolution) | **SUCCESS.** Easily analyzable via the Files API or YouTube URL. | ≈ 108,000 tokens (6 mins × 300 tokens/s) – Uses only ≈ 10% of the 1M token window. |
| **ChatGPT (GPT-4o / GPT-4V)** | Indirect / Image count-based | **SUCCESS (Indirect).** Must be uploaded as a full Transcript and/or a few Screenshots. Cannot be uploaded as a raw video file. | **Text Only:** The transcript will be a low token count (≈ 6,000 tokens) and is easily analyzed in the 128K–200K token window. |
| **Claude (Opus / Sonnet)** | Indirect / Token-based (Extremely High) | **SUCCESS (Indirect).** Must be uploaded as a full Transcript. The long context window handles this text easily. | **Text Only:** The transcript is very small relative to Claude’s huge context window (200K–1M tokens), making analysis fast and easy. |
+
+
+ * **Runway ML** - https://runwayml.com
+ * **Invideo AI** - https://invideo.io/ai
+ * **Pika Labs** -https://pika.art
+ * **Kapwing AI** -https://www.kapwing.com/ai
+ * **Canva AI Video** -https://www.canva.com/ai-video
+ * **Genially** - https://www.genially.com
+
+
+ ## **13\. Some API options**
+
+ * **Shotstack **– A cloud-video-editing API: you send JSON to control timeline, clips, overlays, text, etc. Shotstack+1
+ * **Example**: their “Add Text to Video” guide shows exactly how to programmatically add text overlays. Shotstack
+ * **Creatomate** – Another API for automated video/image generation & editing via templates. creatomate.com+1
+ * **Banuba** – Offers an AI video editing SDK/API (for mobile/web) that supports features like text, overlays, filters. banuba.com
+ * **Cloudinary** – Has a “Video Editing API” feature that includes adding text overlays and other transformation capabilities. Cloudinary+1
+
+ ## **14\. What you should check / consider**
+
+ * Does the API allow text overlay at a specific time, position, animation etc? (Yes — Shotstack has a guide.)
+ * Does it support your required video formats, size (for social media maybe 9:16) and output quality?
+ * Are there template / automation capabilities (useful if you’ll generate multiple videos)
+ * What’s the pricing & scalability (if you’ll generate many videos)
+ * Does the API require that media is uploaded to their cloud or can you point to external sources?
+ * **Integration ease:** languages supported (Node.js, Python etc)
+
+ ## **15\. The best AI video editing software**
+
+ * **Google Veo ** - for end-to-end video creation
+ * **Runway** - for generative AI video with advanced tools
+ * **Sora** - for community-driven inspiration and remixing
+ * **Descript** - for editing video by editing the script
+ * **Wondershare Filmora** - for polishing video with AI tools
+ * **Capsule** - for simplifying video production workflows with AI
+ * **invideo AI** - for social media videos
+ * **Peech** - for content marketing teams
+ * **Synthesia** - for using digital avatars
+ * **HeyGen** - for interactive avatars
+ * **Vyond** - for animated character videos from a prompt
+ * **revid.ai** - for AI-powered templates
+ * **Luma Dream Machine** - for brainstorming with AI
+ * **LTX Studio** - for extreme creative control
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9