feat: AI Video summerization and subtitle generation with Lingo (Closes #1761) (#1891)

* initial setup

* backend code

* updated env and requirements due to some junk files

* fronted setup with lingo with some vibe code

* Read me update

* added lingo sdk in backend + translated text

* updated requirements

* lang send from frontend

* ui update

* readme update

* gitignore update

* change set added

* signed commit

* ai comments resolve

* added vue18-n

* issue fix

* remove change set file and rerun empty file

---------

Co-authored-by: Sumit Saurabh <62152915+sumitsaurabh927@users.noreply.github.com>

Author: longkhan786

.changeset/petite-files-allow.md

@@ -0,0 +1,2 @@
+---
+---

community/video-lingo-ai/.gitignore

@@ -0,0 +1,28 @@
+# Logs
+logs
+*.log
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*
+pnpm-debug.log*
+lerna-debug.log*
+
+node_modules
+dist
+dist-ssr
+*.local
+
+# Editor directories and files
+.idea
+.DS_Store
+*.suo
+*.ntvs*
+*.njsproj
+*.sln
+*.sw?
+
+.env
+.venv
+.venv-new
+__pycache__
+.vscode

community/video-lingo-ai/README.md

@@ -0,0 +1,109 @@
+# Video Lingo AI – Demo
+
+## Overview
+
+**Video Lingo AI** is a web application that allows users to:
+
+* Upload a video file.
+* Extract audio and transcribe it using Whisper AI.
+* Optionally generate a summarized version of the transcript.
+* View the transcript with timestamps.
+* Due to limited resources please upload any english spoked video.
+
+
+This tool is built to demonstrate how AI can help automatically understand, summarize, and localize video content.
+
+---
+
+## Features Highlighted
+
+* **Video Processing & Transcription:** Uses Whisper AI to convert spoken content into text segments with timestamps.
+* **Summarization:** Generates concise summaries of video content (via a language model).
+* **Dynamic Transcript Display:** Shows each segment with start/end times for easy navigation.
+* **Multi-language Support :** Designed to translate transcript and summary into other languages, demonstrating the power of Lingo.dev for localization.
+* **Interactive Vue3 Frontend:** Drag-and-drop video uploads, toggles for summarization, and responsive results display.
+
+---
+
+## Tech Stack
+
+* **Backend:** FastAPI, Python 3.10+, Whisper AI, OpenAI API (for summarization & translation).
+* **Frontend:** Vue 3, Composition API, Axios for API requests.
+* **Other Libraries:** Pydantic, Requests, pathlib, shutil.
+
+---
+
+## How to Run Locally
+
+1. **Clone the repository:**
+
+```bash
+git clone <your-repo-url>
+cd video-lingo-ai/api
+```
+
+2. **Create and activate a virtual environment:**
+
+```bash
+python -m venv .venv
+source .venv/bin/activate   # Linux/Mac
+.venv\Scripts\activate      # Windows
+```
+
+3. **Install dependencies:**
+
+```bash
+pip install -r requirements.txt
+```
+
+4. **Set environment variables:**
+
+Create a `.env` file 
+
+```env
+GROQ_API_KEY=<your_openai_api_key>
+LINGODOTDEV_API_KEY=<your_lingo_api_key>
+```
+
+5. **Run the backend:**
+
+```bash
+uvicorn src.main:app --reload
+```
+
+6. **Run the frontend:**
+
+```bash
+npm install
+npm run dev
+```
+
+7. **Access the app in your browser:**
+
+```
+http://localhost:5173
+```
+
+* Drag and drop a video file to upload.
+* Toggle summarization if needed.
+* View the transcript and summary in English/Hindi.
+
+---
+
+## Lingo.dev Integration
+
+The app is designed to highlight Lingo.dev features:
+
+* Translating dynamic content such as transcripts and summaries.
+* Supporting multiple languages for global accessibility.
+* Easy integration with web apps for localization workflows.
+
+---
+
+## Notes
+
+* The summarization feature may take a few seconds depending on video length.
+
+Thank you for Organizing this
+
+---

community/video-lingo-ai/api/requirements.txt

@@ -0,0 +1,51 @@
+annotated-doc==0.0.4
+annotated-types==0.7.0
+anyio==4.12.1
+av==16.1.0
+certifi==2026.1.4
+charset-normalizer==3.4.4
+click==8.3.1
+coloredlogs==15.0.1
+ctranslate2==4.6.3
+distro==1.9.0
+dotenv==0.9.9
+exceptiongroup==1.3.1
+fastapi==0.128.0
+faster-whisper==1.2.1
+filelock==3.20.3
+flatbuffers==25.12.19
+fsspec==2026.1.0
+groq==1.0.0
+h11==0.16.0
+hf-xet==1.2.0
+httpcore==1.0.9
+httpx==0.28.1
+huggingface-hub==1.3.3
+humanfriendly==10.0
+idna==3.11
+lingodotdev==1.3.0
+mpmath==1.3.0
+nanoid==2.0.0
+numpy==2.2.6
+onnxruntime==1.23.2
+packaging==26.0
+pip==22.0.2
+protobuf==6.33.4
+pydantic==2.12.5
+pydantic-core==2.41.5
+python-dotenv==1.2.1
+python-multipart==0.0.21
+pyyaml==6.0.3
+requests==2.32.5
+setuptools==59.6.0
+shellingham==1.5.4
+sniffio==1.3.1
+starlette==0.50.0
+sympy==1.14.0
+tokenizers==0.22.2
+tqdm==4.67.1
+typer-slim==0.21.1
+typing-extensions==4.15.0
+typing-inspection==0.4.2
+urllib3==2.6.3
+uvicorn==0.40.0

community/video-lingo-ai/api/src/__init__.py

@@ -0,0 +1,62 @@
+from dotenv import load_dotenv
+load_dotenv()
+
+from fastapi import FastAPI, UploadFile, File
+from fastapi.responses import JSONResponse
+from pathlib import Path
+import shutil
+import tempfile
+
+from .models import whisper, generate_text 
+from .utils.utils import extract_audio, translate_text_with_lingo       
+from fastapi.middleware.cors import CORSMiddleware
+
+app = FastAPI(title="Video Lingo AI API")
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+# for demo project above i have allowed all origins. 
+
+UPLOAD_DIR = Path(tempfile.gettempdir()) / "video_lingo_ai"
+UPLOAD_DIR.mkdir(parents=True, exist_ok=True)
+
+@app.post("/process-video")
+async def process_video(
+    file: UploadFile = File(...),
+    issummarize: bool = False,
+    lang: str = "en"
+):
+    video_path = UPLOAD_DIR / file.filename
+    with open(video_path, "wb") as f:
+        shutil.copyfileobj(file.file, f)
+
+    audio_file = extract_audio(str(video_path))
+
+    segments, _info = whisper.transcribe(str(audio_file))
+    segments = list(segments)
+
+    full_text = " ".join(seg.text for seg in segments)
+
+    if issummarize:
+        summary = generate_text(f"Summarize this video:\n{full_text}")
+        if lang != "en":
+            summary = await translate_text_with_lingo(summary, lang)
+        return JSONResponse({"summary": summary})
+
+    result = []
+    for seg in segments:
+        text = seg.text
+        if lang != "en":
+            text = await translate_text_with_lingo(text, lang)
+
+        result.append({
+            "start": seg.start,
+            "end": seg.end,
+            "text": text
+        })
+
+    return JSONResponse({"segments": result})

community/video-lingo-ai/api/src/main.py

@@ -0,0 +1,52 @@
+from faster_whisper import WhisperModel
+import os
+from groq import Groq
+import re
+
+whisper = WhisperModel(
+    "tiny",  
+    device="cpu", # i dont have a gpu so it can be slow 
+    compute_type="int8"
+)
+
+client = Groq(
+    api_key=os.getenv("GROQ_API_KEY")
+)
+
+
+def extract_final_answer(text: str) -> str:
+    if "FINAL_ANSWER:" in text:
+        text = text.split("FINAL_ANSWER:", 1)[1]
+
+    text = re.sub(r"<think>.*?</think>", "", text, flags=re.DOTALL)
+
+    text = text.strip().strip('"').strip("'")
+    text = re.sub(r"^[^A-Za-z0-9]+", "", text)
+
+    return text.strip()
+
+def generate_text(prompt: str) -> str:
+    response = client.chat.completions.create(
+        model="qwen/qwen3-32b",
+        messages=[
+            {
+                "role": "user",
+                "content": (
+                    "Summarize the following transcript.\n\n"
+                    "STRICT RULES:\n"
+                    "- No reasoning\n"
+                    "- No explanations\n"
+                    "- No analysis\n"
+                    "- Output ONLY the final summary\n"
+                    "- Start with: FINAL_ANSWER:\n\n"
+                    + prompt
+                )
+            }
+        ],
+        temperature=0.2,
+        max_tokens=512,
+    )
+
+    raw = response.choices[0].message.content
+    return extract_final_answer(raw)
+

community/video-lingo-ai/api/src/models.py

@@ -0,0 +1,22 @@
+import subprocess
+import os
+from lingodotdev import LingoDotDevEngine
+
+def extract_audio(video_path, output_audio="temp_audio.wav"):
+
+    command = ["ffmpeg", "-y", "-i", video_path, "-ac", "1", "-ar", "16000", output_audio]
+    subprocess.run(command, stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)
+
+    if not os.path.exists(output_audio):
+        raise RuntimeError("Audio extraction failed")
+
+    return output_audio
+
+
+async def translate_text_with_lingo(text: str, target_lang: str) -> str:
+    result = await LingoDotDevEngine.quick_translate(
+        text,
+        api_key=os.getenv("LINGODOTDEV_API_KEY"),
+        target_locale=target_lang
+    )
+    return result

community/video-lingo-ai/api/src/utils/utils.py

@@ -0,0 +1,15 @@
+{
+  "version": "1.11",
+  "locale": {
+    "source": "en",
+    "targets": ["hi", "es", "fr"]
+  },
+  "buckets": {
+    "json": {
+      "include": [
+        "i18n/[locale].json"
+      ]
+    }
+  },
+  "$schema": "https://lingo.dev/schema/i18n.json"
+}