How to use OpenAI’s Whisper for better speech-enabled (audio) search — OpenAI’s Whisper is a new state-of-the-art (SotA) model in speech-to-text. It can almost flawlessly transcribe speech across dozens of languages and even handle poor audio quality or excessive background noise. The domain of spoken word has always been somewhat out of reach for ML use cases. Whisper changes that for…

Enhancing GIF discovery using NLP and semantic search — Vector search powers some of the most popular services in the world. It serves your Google results, delivers the best podcasts on Spotify, and accounts for at least 35% of consumer purchases on Amazon [1][2]. In this article, we will use vector search applied to language, called semantic search, to…

Learning doesn’t need to be so hard — The modern world rewards those that can learn and adapt quickly. Both financially and in your own personal fulfillment. Learning new skills keeps life interesting and fresh, but it isn’t easy to do. Continuous learning can fry your brain very quickly. Yet, there’s nothing better than expanding the scope of…

Learn how to translate between text to image and back again with CLIP and vector embeddings — After a few short years of life, children can fathom the concepts behind simple words and connect them to related images. They can identify the connection between shapes and textures of the physical world to the abstract symbols of written language. It’s something we take for granted. Very few (if…

Making diffusers models like DALL-E 2 and Imagen more accessible than ever — Hugging Face, the creators of the transformers library, have just released a whole new library for building diffuser models. If you’re not sure what diffuser models are, think of them as the not-so-secret secret behind some of the most talked about AI models this year. Those insanely beautiful, creative, almost…

Easy guide to Hugging Face and Sentence-Transformers on M1 with GPU-acceleration — Prerequisites MacOS 12.4+ Anaconda distribution for M1 (find at bottom of page here) Python Environment We first setup a new Python environment that uses osx-arm64 and install the latest version(s) of PyTorch. Open terminal Create new environment with conda create -n ml python=3.9 anaconda [Alternative]: If using the non-M1 distribution of Anaconda use…

How do the new M1 chips perform with the new PyTorch update? — The release of M1 Macs in November 2020 marked a significant step up in the processing power of Apple machines [1]. Unfortunately, these new features were not integrated into PyTorch until now. Today’s deep learning models owe a great deal of their exponential performance gains to ever increasing model sizes…

Practical resources and advice from a self-taught NLP engineer — Knowing what to learn is one of the hardest parts about self-learning. Imagine being thrown into the wilderness and being told to find a specific landmark. Without a map you will end up wandering to wilderness with no better option than taking one step after another. I spent a long…

How to build *everything* with Streamlit — Streamlit is well-loved as a pure Python solution to building web apps. It makes building dynamic, data-centric apps incredibly easy. And the learning curve is practically non-existent.