In Part 1 I found that my Malayalam speech model’s test set leaked into its training data. In Part 2 I re-tested the model on clean data only, and promised a better v2. This post reports two things. First, v2 is trained, and on the same clean test set it cuts word errors by 12
Part 3: Retraining My Malayalam ASR Model, and a Bug in My Own Evaluation Read More »
Correction (11 October 2026): The v1 numbers in this post came from an evaluation notebook that capped transcripts at 225 tokens. Malayalam needs about 25 Whisper tokens per second of speech, so clips longer than about 9 seconds were cut off and the missing words counted as errors. With the cap fixed, v1 scores 43.5%
I Re-Tested My Malayalam ASR Model on Clean Data — Part 2 Read More »
INTELEVO EP 05 – Complete Tutorial for AI & Data Science Introduction Imagine you’re a teacher who needs to print grade reports for 150 students. Without loops, you’d write 150 individual print statements. With loops? Just five lines of code. That’s the power of iteration. Loops are one of the most powerful features in programming.
Loops in Python: The Complete Guide to for, while, and Loop Control Read More »
Softmax activation function is a commonly used activation function in deep learning. In this tutorial, let us see what it is, why it is needed, and how it is used. Softmax activation function is commonly used in the output layer of classifier neural network. What is a classifier? Lets start with understanding what a classifier
Softmax Activation Function Read More »
Some people walk into a room, and all eyes follow. Their energy feels effortless, drawing others in with ease. This pull isn’t just about looks or words—it’s the Charisma Quotient in action. A mix of confidence, intelligence, and charm makes someone truly magnetic. Attraction isn’t about being the loudest or the most perfect. It’s about
The Gravity of Charm: Where Charisma Quotient Weaves Attraction Read More »
