InSciforcebySciforce·Nov 2How Voice AI Automates Customer Conversations Across HospitalityMost calls to a restaurant or hotel end with someone updating a record. A table is held for Friday night, or a guest’s checkout moves to…
Esteban Gómez·Oct 31Tracking machine learning experiments with NuBoardWhen I train a model, I usually run many variations of it: a few learning rates, a couple of seeds, an ablation or two. TensorBoard shows…
Thanabordee Noun·Sep 25LAYA BUT AUDIO AS INPUTเราจะให้ Laya รับเสียงแล้วตัดสินใจได้เลยไหม โดยไม่ต้องถอดเสียงเป็นข้อความก่อน
Ashwini Aher·Sep 9Speech-Based Wireless Channel Quality Classification Using Machine Learning, SQL and EDAIn wireless communication, channel conditions directly affect the quality of a received signal. Factors such as noise and fading can…
Hemang Baldha·Aug 13Building a Deep Learning-Based Voice Activity Detection Model from ScratchBuilding a Deep Learning-Based Voice Activity Detection Model from Scratch
InTowards AIbyAnil Mandra·Aug 10What I Got Wrong Fine-Tuning F5-TTS for a New Writing SystemFive failures that cost GPU hours, and the measurements that found them.
InTowards AIbyAnil Mandra·Jul 13Building a Production Audio Separation API with Meta’s DemucsHow to turn a research model into a deployable service for vocal isolation, podcast cleanup, and content localization.
Vatsal Jha·Jul 3One-Class Learning for Audio Deepfake DetectionWhy Teaching a Model What Real Sounds Like Is Better Than Teaching What Fake Sounds Like
Riya Kaulwar·May 22Why My Parkinson’s Speech Model Failed with a 0.71 AUC and What Fixed ItAn exploration of why early biomedical speech models fail to generalize, and how disciplined data splitting, augmentation, and…
Aiza Khurram·Apr 26Speech Processing (Load, Visualization and features extraction) using Python(librosa)Every time you speak into a microphone, your voice is converted from continuous air pressure (analog signal) into discrete numerical values…
