The latest articles on DEV Community by Samarth Murade (@samarthmurade3105). https://dev.to/samarthmurade3105 https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fthepracticaldev.s3.amazonaws.com%2Fi%2F99mvlsfu5tfj9m7ku25d.png https://dev.to/samarthmurade3105 en Samarth Murade Mon, 05 Oct 2026 02:22:06 +0000 https://dev.to/samarthmurade3105/lecture-lens-clearer-audio-transcripts-and-summaries-for-quiet-lectures-1bm https://dev.to/samarthmurade3105/lecture-lens-clearer-audio-transcripts-and-summaries-for-quiet-lectures-1bm <h2> What I built </h2> <p>Lecture Lens makes a quiet, noisy class recording louder and clearer, and turns it into a Hinglish transcript, a summary, key points and revision questions. It runs entirely on my laptop.</p> <h2> Who I built it for and the problem </h2> <p>My sister. She sits close to the teacher and records every lecture with a voice recorder, because she cannot follow everything in class the first time. She then listens to the recording again at home. She had two problems:</p> <ol> <li> <strong>The audio was too quiet.</strong> The recorder picked up the teacher's voice very faintly, so even at home it was hard to hear.</li> <li> <strong>A recording is slow to revise from.</strong> To get back to one explanation, she had to listen to the whole lecture again. She needed the lecture as text: a transcript to read, and a short summary with the key points.</li> </ol> <p>So Lecture Lens does both. It makes the audio clearer and louder, and it turns the lecture into a transcript, a summary, key points and revision questions.</p> <h2> Demo </h2> <p> <iframe src="https://www.youtube.com/embed/1PhxMjQOOC0" width="710" height="399"> </iframe> </p> <h2> Code </h2> <p><a href="https://github.com/samarth-murade/lecture-lens" rel="noopener noreferrer">https://github.com/samarth-murade/lecture-lens</a> (MIT license)</p> <h2> How I used open-source AI </h2> <ul> <li> <strong>Speech to text:</strong> faster-whisper with a Hinglish fine-tuned Whisper model (a community CTranslate2 conversion), so Hindi and English speech come out in English letters.</li> <li> <strong>Notes:</strong> Gemma 4 (through Ollama) writes the summary, key points and 5 revision questions.</li> <li> <strong>Audio:</strong> ffmpeg filters (noise reduction, loudness normalization). This part is not AI.</li> <li> <strong>Interface:</strong> Streamlit.</li> </ul> <h2> Why open innovation mattered here </h2> <p>Classroom recordings are private, and a laptop at home may have no reliable internet. Because the models are open, everything runs on my own laptop: nothing is uploaded, there is no subscription, and the models can be swapped for better ones later.</p> <h2> Limits (honest notes) </h2> <ul> <li>It runs on an ordinary laptop CPU with no GPU, so long lectures are slow.</li> <li>Transcript quality depends on the recording: good on clear audio, weaker on noisy parts.</li> <li>Notes are written by AI and can contain mistakes.</li> </ul> <h2> Built with </h2> <p>Python, Streamlit, ffmpeg, faster-whisper, Gemma 4 and Ollama. I built it over a weekend with the help of an AI coding assistant (Claude) and tested it on a real classroom recording.</p> devchallenge weekendchallenge hf26challenge Samarth Murade Thu, 01 Oct 2026 14:02:38 +0000 https://dev.to/samarthmurade3105/introduction-3fmh https://dev.to/samarthmurade3105/introduction-3fmh <p>Hello, I am new to open source,glad to be part of Hactoberfest 2026!</p>