Natural Language Processing news and analysis on Web Pulse: 3969 curated articles tracking the latest Natural Language Processing developments, tools, and research, updated continuously from vetted sources.
Google updated Gemini for Mac with a system-wide voice dictation mode activated by long-pressing the Fn key, allowing users to speak naturally from any window. The update also introduces an opt-in screen-aware reasoning …
A design pattern for local AI workflows, illustrated through a UDRP-inspired domain-risk pipeline, shows how existing industry processes can serve as blueprints for AI agent workflows. The pipeline breaks domain name tra…
Google Cloud announced the preview of Looker Agentic Workflows, a new capability in Looker that automates metric monitoring and root-cause analysis using intelligent background agents. The feature allows users to set up …
Google is rolling out advanced voice control for Gemini for macOS, enabling intelligent dictation and context-aware commands. The update, announced at I/O 2026, lets users speak naturally into any window, with features l…
Google announced Wednesday that its Gemini app for MacOS is adding intelligent dictation and a Screen Awareness mode, allowing the AI to see active windows and execute tasks like turning highlighted notes into an executi…
Google's Gemini desktop app on macOS now lets users press and hold the fn key to issue voice commands from any app, with the AI able to understand on-screen content to rewrite documents, polish emails, and summarize note…
Encore AI, a startup that analyzes customer interactions to train AI voice agents, has raised $30 million in a Series A round led by Team8. Founded in 2022 by CEO Dvir Ginzburg, the company rebranded from Insait IO and n…
Daydream, an AI-powered fashion search platform, today launches Powered by Daydream, embedding natural-language search into brand websites. Staud, Alice + Olivia, Couper, Cult Mia, and Hampden Clothing are live in the in…
IOS 27, launching this fall, introduces three major features: a rebuilt Siri AI that provides informed answers and performs more actions, AI-powered automatic proofreading including grammar checks, and a search overhaul …
A self-taught AI engineer and amateur linguist claimed a breakthrough in deciphering Linear A, the undeciphered writing system of the Bronze Age Minoan civilization, by using AI-built programming scripts to test a hypoth…
OpenAI released GPT Transcribe and GPT Live Transcribe, two new speech recognition models available through its API, with GPT Transcribe achieving a 3.31 percent word error rate on the AA-WER benchmark, a 0.7 percentage …
Echologue, a private AI voice journal built by an anonymous developer, stores all data on-device and uses zero-data-retention endpoints for AI inference, allowing users to ask natural-language questions about their entri…
ElevenLabs has expanded its ElevenAgents platform to support omnichannel deployment across phone, web, WhatsApp, SMS, Telegram, Slack, and ticketing systems including Zendesk, Intercom, and Freshdesk. The update enables …
A year-long study of 15.7 million AI Mode citations across 148 industries by SEO tool Pillarbase reveals that Google's AI Mode cites specific passages rather than entire pages, with 47.7% of citations being scroll-to-tex…
Oye Collective has built a local PII detection feature for coding agent transcripts that scans for secrets without sending data to the cloud. The system combines regex for structured tokens like API keys with a quantized…
Trustample, an AI transcription platform, implemented a simple arithmetic check to verify its own output after a user uploaded a Spanish recording that was processed as English, producing a nearly empty transcript withou…
HBO Max launches a TikTok-style vertical short-form feed called 'HBO Max Shorts' for iOS users in the US and an experimental AI search feature 'Ask HBO Max' for Android users aged 18 and older, parent company Warner Bros…
Google announced that Gemini in Google Docs can now read, understand, and take action based on comments, including summarizing comments, identifying key themes, responding to threads, and suggesting changes. The rollout …
Liquid AI released two open-weight bidirectional encoders, LFM2.5-Encoder-230M and LFM2.5-Encoder-350M, built on the LFM2 hybrid backbone with an 8,192-token context. The 350M model achieves a 17-task mean of 81.02, rank…
A Communications Psychology study published April 28 analyzed 3,366 dream and waking-experience reports from 207 adults, plus 351 dream reports from 80 adults during Italy's first COVID-19 lockdown, finding associations …