🎙️ Podcast Summarizer & YouTube Suggester

A Streamlit-based web application that allows users to upload a podcast audio file or provide a YouTube link to get a summary, keyword extraction, and relevant podcast suggestions. The project leverages AI-powered speech-to-text, NLP, and summarization models to deliver concise podcast insights.

Features

🎧 Podcast & YouTube Audio Processing

Upload an MP3 file or provide a YouTube link.

Uses yt_dlp to download and convert YouTube audio.

📝 AI-Based Transcription

Uses OpenAI’s Whisper model to transcribe the audio.

📜 Summarization

Summarizes the podcast using the Google Long-T5 model.

🔍 Keyword Extraction & Podcast Recommendations

Extracts key topics using spaCy. Searches for related long-form podcasts on YouTube using the YouTube API.

🔊 Text-to-Speech (TTS) Playback

Uses Deepgram TTS API to convert the summary into speech. Plays the summary audio using ffplay. 🎨 User-Friendly Web Interface

Built using Streamlit for an interactive UI.

🛠️ Installation & Setup

1️⃣ Clone the Repository
git clone https://github.com/absarraashid3/PodcastSummarizerAndYoutubeSuggester.git
cd PodcastSummarizerAndYoutubeSuggester

2️⃣ Install Dependencies
Ensure you have Python 3.8+ installed. Then run:
pip install -r requirements.txt

3️⃣ Install SpaCy Model
python3 -m spacy download en_core_web_sm

4️⃣ Set API Keys
Create a .env file in the project root and add:
YOUTUBE_API_KEY=your_youtube_api_key_here
DEEPGRAM_API_KEY=your_deepgram_api_key_here

5️⃣ Run the Application
python -m streamlit run PodcastSummarizerAndYoutubeSuggester.py

🖥️ Usage

1️⃣ Launch the app using the command above.

2️⃣ Select an input type:

Upload an MP3 file. Enter a YouTube URL.

3️⃣ Get the transcript & summary.

4️⃣ View suggested podcasts based on extracted keywords.

5️⃣ Listen to the summary via text-to-speech.

📦 Project Structure

📂 PodcastSummarizerAndYoutubeSuggester/
├── 📜 requirements.txt
├── 📜 PodcastSummarizerAndYoutubeSuggester.py
├── 📂 downloads/ (stores downloaded audio files)
├── 📂 models/ (stores NLP models if needed)
└── 📜 README.md

🛠️ Tech Stack

Python
Streamlit (Frontend UI)
Whisper AI (Speech-to-Text)
spaCy (Keyword Extraction)
Hugging Face Transformers (Summarization Model)
YouTube API (Podcast Search)
Deepgram API (Text-to-Speech)
pydub (Audio Processing)

🔥 Future Enhancements

✅ Improve multi-language support for transcriptions.
✅ Add user preferences for podcast recommendations.
✅ Enhance UI/UX with better design.

📩 Contact

For queries, reach out via:

Email: absarrashid3@gmail.com