LingoLens 🎤: Speak, Transcribe, Translate

This is a submission for the AssemblyAI Challenge: Sophisticated Speech-to-Text && Really Rad Real-Time && No More Monkey Business.

---

What I Built

LingoLens 🎙️ is a web application designed to revolutionize how we process and analyze audio content.

Powered by AssemblyAI's LEMUR API and Gemini for enhanced AI processing, it allows users to transcribe, translate, and gain insights from audio files in real-time.

With an interactive, animated UI built using Next.js, TailwindCSS, and Framer Motion, LingoLens makes audio data accessible and actionable across different languages and contexts.

---

Key Features:

• 🎙️ Real-Time Speech-to-Text: Converts speech into text in multiple languages.
• 🌍 Language Translation: Translates the transcriptions into various languages.
• 🗣️ Speaker Diarization: Identifies and labels speakers in the conversation.
• 📊 Audio Analytics: Provides sentiment analysis, keyword extraction, and summarized insights.
• 💫 Interactive UI: Beautiful animations and smooth user transitions powered by Framer Motion.
• 📂 Export Options: Download or share transcriptions and analytics.

---

Demo

🎥 YouTube Demo: ↘️

🦄 Live Demo: LingoLens Live in Vercel

😻 GitHub Repository: LingoLens Codebase

---

Screenshots 📸

Here’s a glimpse of the app in action:

1. 🏠 Home Page
   
   

2. 📂 Upload Audio Page
   
   

3. 📜 Results Page
   
   

✅ Select language to translate

✅ Translating to Chinese

✅ Translated to Chinese

✅ Generated Blog from transcript

✅ Generated Social Post (summary) from transcript

✅ Ask any Question related to transcript

✅ Get instant answer

✅ Save this page in history for future reference

4.🪄History

5.🅰️About LingoLens

6.📞Contact Me

---

Journey

🎤 Universal-2, AssemblyAI’s Speech-to-Text model, is at the core of this application, enabling efficient and accurate transcriptions. Here’s how the LEMUR API powers the features:

1. ✍️ Transcription: Real-time conversion of audio to text, supporting multiple languages.
2. 🌐 Translation: Transcriptions are sent to AssemblyAI for language translation, making the app globally accessible.
3. 🧑‍🤝‍🧑 Speaker Diarization: Identifies multiple speakers within the audio and segments the transcription accordingly.
4. 📈 Audio Analytics: Sentiment analysis and keyword extraction from the transcriptions are powered by AssemblyAI’s advanced processing.

---

Additional Tools and Prompts

• ⚡ Qualified for the Really Rad Real-Time prompt by enabling live transcription and analysis in real-time.
• ✨ Integrated Framer Motion to add delightful animations to the UI for better user interaction.

---

Team Submissions

👨‍💻 This is a solo submission by Aniruddha Adak. You can find the code on my GitHub repository.

---

Thanks for reading! 😊