AssemblyAI has made it much easier to distinguish speakers and determine what they spoke in a conversation, resulting in:
13% more accurate transcript than previous versions.
85.4% reduction in speaker count errors.
5 new languages (total 16 supported languages).
A demo is shown below.
First, import the package, set the API key, and transcribe the file while setting speaker_labels parameter to True:
Next, print the speaker labels:
AssemblyAI’s speech-to-text models rank top across all major industry benchmarks. You can transcribe 1 hour of audio in ~35 seconds at an industry-leading accuracy of 92.5% (for English).
Histograms are quite common in data analysis and visualization. Yet, they can be highly misleading at times.
Why?
To begin, a histogram represents an aggregation of one-dimensional data points based on a specific bin width:
This means that setting different bin widths on the same dataset can generate entirely different histograms.
This is evident from the image below:
Altering the bin width changes the type of histogram created
As shown above, each histogram conveys a different story, even though the underlying data is the same.
Thus, solely looking at a histogram to understand the data distribution may lead to incorrect or misleading conclusions.
Here, the takeaway is not that histograms should not be used. Instead, it is that Whenever you generate any summary statistic, you lose essential information.
In our case, every bin of a histogram also represents a summary statistic — an aggregated count.
Histograms depict a summary statistic
And whenever you generate any summary statistic, you lose essential information.
Thus, it is always important to look at the underlying data distribution.
For instance, to understand the data distribution, I prefer a violin (or KDE) plot. This gives me better clarity of data distribution over a histogram.
Left: KDE plot; Right: Violin plot
Visualizing density provides more information and clarity about the data distribution than a histogram.
👉 Over to you: What other measures do you take when using summary statistics?
Consider two functions that take a decent amount of time to execute and return a boolean:
long_function
longer_function
We want to run a conditional if one of them returns True. An optimal way to do this is by shifting the function call in the if statements.
This way, if long_process() returns True, longer_process() will not be executed because of the way OR works. This reduces run-time.
A similar optimization can be achieved if we intend to use AND.
THAT'S A WRAP
SPONSOR US
ADVERTISE TO 450k+ Data Professionals
Our newsletter puts your products and services directly in front of an audience that matters — thousands of leaders, senior data scientists, machine learning engineers, data analysts, etc., around the world.
Master Full-stack AI Engineering In today's newsletter: The operating system for AI research labs! Why agent crashes are nothing like database crashes. [Hands-on] Deploy a Qwen 3 Agentic RAG. TODAY'S ISSUE OPEN-SOURCE The operating system for AI research labs! Transformer Lab is an open-source ML platform that orchestrates GPUs across any cloud and runs any training or eval workflow you define: supports LoRA, QLoRA, DPO, ORPO, SIMPO use it from a GUI, CLI, or agent skill. works with MLX,...
Master Full-stack AI Engineering In today's newsletter: Fine-tune any LLM directly from Claude! Speculative decoding in LLMs. tSNE Projections can be misleading. TODAY'S ISSUE fine-tuning Fine-tune any LLM directly from Claude! We built a Hugging Face fine-tuning studio that lets you fine-tune any LLM directly from Claude: The app connects to the HF Hub for model and dataset search. It handles chat template formatting for the training data, and lets you configure LoRA rank, quantization,...
Master Full-stack AI Engineering In today's newsletter: Markov decision processes and value functions in RL. How LLM inference works internally. TODAY'S ISSUE AI engineering Markov decision processes and value functions in RL Last week, we launched a hands-on course series on reinforcement learning. Part 2 is now available, and you can read it here → Reinforcement learning nanodegree part 2 Part 1 gave you the RL interaction loop and the exploration-exploitation tradeoff through bandits, and...
