Update README.md

# AudioSummarizer 🎙️

## Overview
AudioChatTranscriber is a simple web app that transcribes and summarizes audio from multiple sources — local upload, microphone, or YouTube URL.
It uses **faster-whisper** for transcription and **Azure Phi-4-multimodal-instruct** for structured summarization, with a clean **Gradio UI**.

---

## Features
- Upload or record audio, or provide a YouTube/MP3 URL
- Transcribes YouTube with `faster-whisper` + `yt-dlp`
- Summarizes audio/text with **Azure AI**
- Structured outputs: **Summary • Key Details • Insights**
- Works locally or on Hugging Face Spaces with `ffmpeg` support

---

## Quick Start
- Click **Upload**, **Record**, or enter a **YouTube URL**
- Adjust prompts if needed (defaults loaded from `metadata.json`)
- Hit **Summarize** → get instant structured output

---

## Tech Stack
- **Gradio** – User interface
- **faster-whisper** – Transcription
- **yt-dlp + ffmpeg** – YouTube audio extraction
- **Azure Phi-4-multimodal-instruct** – Summarization

---

## License
MIT License. See [LICENSE](./LICENSE).

README.md

@@ -1,15 +1,15 @@
 ---
-title: AudioChatTranscriber
+title: AudioSummarizer
 emoji: 🐠
 header: mini
 colorFrom: green
 colorTo: yellow
 sdk: gradio
-sdk_version: 5.46.0
+sdk_version: 5.46.1
 app_file: app.py
 pinned: false
 license: mit
 short_description: Summarize audio from three sources=File, Microphone & URL
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference