Microsoft Launches Vector Search in Preview, Voice Cloning in General Availability

At its annual Inspire conference, Microsoft announced a number of new AI features headed to Azure, perhaps the most notable of which is Vector Search. From a report: Available in preview through Azure Cognitive search, Vector Search uses machine learning to capture the meaning and context of unstructured data, including images and text, to make search faster. Vectorization, an increasingly popular technique in search, involves converting words or images into vectors, or series of numbers, that encode their meaning -- allowing them to be processed mathematically. Vectors enable machines to structure and make sense of data, enabling them to understand, for example, that words close together in "vector space" -- like "king" and "queen" -- are related and quickly surface them from a database of millions of words.

[...] Rounding out the AI unveilings at Inspire, Microsoft announced the public preview of Real-time Diarization, an AI-driven speech service that can identify which of several people are speaking in real time. The company also announced the general availability of Custom Neural Voice, which taps AI to closely reproduce an actor's voice or create an original synthetic voice. Previously, Custom Neural Voice was in limited access, meaning that customers had to apply and be approved by Microsoft in order to use it.

Deepfake for voice...

[The New Guy 2.0]

This has to be restricted... we don't want a SAPI-like deepfake for voice.

Probably too late to restrict DeepFakeVoice

[davidwr]

I think that genie left the bottle awhile back.

To put it another way, Elvis, or maybe DeepFakeElvis, I can't be sure, they sound so much alike, has left the building.

Nearly none of the results contain my search terms

[Arnonyrnous Covvard]

Can you please stop second-guessing me? Even putting every word in quotes doesn't guarantee anymore that the search results actually have the search terms. Using search engines has become a total crapshoot. I am not looking for millions of results that look to a search AI like what I might have meant. I am looking for the two dozen pages that have all the actual terms I searched for.

Ah...

[Motleypuss]

...Latent space. I'm not sure how this will improve search, where we require deterministic results. LLMs and anything that uses portions of the same algs, IMO, is a great bullshitter but not much use for getting results. Unless what you're looking for is dynamic, free-flowing, evolving and scatterbrained textual companionship, in which case LLMs work great! But please, don't use the algs with search.