Natural Language Processing

Superwhisper released S1-mini, a 462 MB open-weights text normalizer fine-tuned from Qwen/Qwen3-0.6B, which turns raw ASR transcripts into clean written text with 94.8% token accuracy on a held-out set of 7,519 cases. Th…

Vale, an open-source command-line linter for prose, enforces consistent editorial style across documents in 12 formats entirely offline, with no runtime and parallel file linting. Written in Go and MIT-licensed, Vale is …

Google Home is rolling out an August 20 update that improves Spotify command recognition, changes broadcasts to a two-step process, and adds voice control for setting smart light color temperature to specific kelvin valu…

Anthropic has introduced a new output style for its Claude AI model, called 'Plain English,' which instructs the model to lead with the answer, use common vocabulary, and keep responses concise. The style guide, publishe…

Open-source and commercial 'unslop' tools that rewrite AI-generated text to sound human are emerging as a new software category, with at least three MIT-licensed projects on GitHub, a pip install unslop on PyPI, and mark…

A prompt template that assigns a specific persona and constraints turns any large language model into a sharper, more creative writer, according to testing by an unnamed developer. The structure, which pairs a detailed p…

A developer released Claude Deslop, a prompt that rewrites text to remove AI slop and LLM-isms. The tool enforces plain language, bans certain vocabulary and em-dashes, and removes meta commentary and process narration. …

Metric founder and CEO Hrant Davtyan released ArmBench-ASR on August 20, a benchmark evaluating nearly 30 speech recognition systems on 20.7 hours of Armenian audio across five datasets, with Google's Gemini 2.5 Pro lead…

Researchers at the USC Institute for Creative Technologies, in collaboration with the US Army Research Laboratory, won the Best Paper Award at MAGMaR for a study on how robots can communicate observations to human teleop…

Lexiqr, a new pip-installable Python library, resolves tenant-specific jargon to canonical entities deterministically, with typo tolerance and multilingual support. The library, requiring Python 3.10+ and using RapidFuzz…

Blind Egyptian entrepreneur Mark Morad has launched an AI-assisted app that narrates the visual world for people with vision loss, supporting Arabic and English and processing some tasks on-device. The app, developed in …

DoctPro has implemented a chatbot on its website to provide first-level support, handling common queries about the application, job opportunities, and partnerships. The chatbot uses API integration to fetch real-time inf…

NetDocuments announced two new AI capabilities for its document management platform: Tabular Review, which converts document collections into structured tables for due diligence and contract audits, and Legal Authorities…

Pine AI's 1.2B-parameter speech recognition model tops the τ³-Voice leaderboard with a score of 75, achieving significant gains on accented speech, medical dictation, and code-switching, but only marginal improvement on …

Pine AI's model achieved a 75.4% score on the τ³-Voice Leaderboard, setting a new state-of-the-art (SoTA) for text agents resolving banking customer-service tasks over a ~700-document knowledge base, as published in τ-kn…

OSINTech published its weekly curated roundup of OSINT, digital investigation, and security research tools for the week of 14–20 August 2026, highlighting projects including Registry Lookup, Fleetbase, Semtools, and a lo…

Talecut, an AI-powered video editing tool, converts movies and TV episodes into vertical micro-episodes for platforms like TikTok, Reels, and Shorts, using AI scene analysis to track speakers and generate captions. The t…

Aslema, a system developed for the NADI 2026 Shared Task 5, achieved 86.8% intent accuracy and 34.7 WER on the devtest split, and ranked 1st in slot filling (59.5 CoER) and 4th in intent recognition (66.1% accuracy) on t…

A new arXiv study (2608.18158v1) evaluating LLMs as data quality annotators found that on the Abt Buy entity matching benchmark (2,194 labeled pairs), a rule-based baseline (F1=0.950) performed about as well as LLM zero-…

A scoping review published on arXiv (2608.18135v1) found that artificial intelligence technologies integrated into rural medication management reduce prescribing and transcription errors by 34% to 80%, while facing barri…