# forck.live — 26 September – 2 October 2022 > AI lab announcements, news reports and podcasts. forck.live follows what companies announce about their own products. Press may corroborate an official story; it never leads and never appears as an item of its own. A missing date stays missing. A title is the publisher's words — where a company does not publish in English, an English rendering is listed separately as `Title in English` and named as this site's translation, never as theirs. This file is one issue of the weekly archive: every confirmed first-party announcement of ISO week 2022-W39, 26 September – 2 October 2022, newest first. The window never moves, so this URL always names the same seven days — but a post published inside them and detected later is added when it arrives, so a closed week can still gain entries. Every entry is one first-party announcement, and carries the same eight fields in this order before any others: Company · Announced · Category · Coverage · Announcement · Group · Source · Record `Source` is the company's own page the announcement was traced to and `Record` is this site's page for it. Both are safe to cite. `Coverage` is a number of outlets, or `not counted` — which means this record reached no count of its own, never that nobody wrote about it. `Category` is this tracker's own analysis, and is authoritative wherever it is stated. `not stated` means either that nothing has analysed the post yet or that the analysis placed it outside this taxonomy; neither is a verdict about the announcement. `Announcement` is whether the post is news of something the company did, rather than a how-to, a customer story, a changelog listing or a staffing note. `Group` is which of this record's groups the entry falls in, carried on the entry itself so it holds where the groups are headings and where they are not. One of: `models`, `covered`, `announcements`, `routine`. Anything after `Record` is held detail that only some announcements carry. - This week's page: https://forck.live/week/2022-W39 - The current issue, and the catalogue of the other weeks: https://forck.live/briefing.md - Catalogue: https://forck.live/llms.txt ## Announcements ### LG AI Research — LG Demonstrates Competitiveness at 'Interspeech 2022', the Premier Conference in the Speech Processing - Company: LG AI Research (lgresearch.ai) - Announced: 2022-09-29 - Category: not stated - Coverage: not counted - Announcement: no - Group: routine - Source: https://www.lgresearch.ai/news/view?seq=255 - Record: https://forck.live/items/11063-lg-demonstrates-competitiveness-at-interspeech-2022-the-premier-conference-in - Subject: EXAONE At Interspeech 2022, LG AI Research demonstrated YouTube news video speech-to-text (STT) technology that extracts keywords while converting audio to text in real-time. The research also introduced technology to capture the essence of voice consultations, categorizing content and emotional states. LG Electronics presented papers on user-defined wake-up-word recognition and speaker identification. ### LG AI Research — LG Demonstrates Competitiveness at 'Interspeech 2022', the Premier Conference in the Speech Processing - Company: LG AI Research (lgresearch.ai) - Announced: 2022-09-29 - Category: research-paper - Coverage: not counted - Announcement: yes - Group: announcements - Source: https://www.lgresearch.ai/blog/view?seq=255 - Record: https://forck.live/items/4931-lg-demonstrates-competitiveness-at-interspeech-2022-the-premier-conference-in - Subject: EXAONE LG AI Research, LG Electronics, and LG U+ demonstrated voice recognition technologies and presented three research papers at Interspeech 2022, including STT, voice-controlled appliances, and speaker recognition. ### Replit — Fluid layout customization with Splits - Company: Replit (replit.com) - Announced: 2022-09-28T19:06:40+00:00 - Category: capability-change - Coverage: not counted - Announcement: yes - Group: announcements - Source: https://replit.com/blog/splits - Record: https://forck.live/items/18065-fluid-layout-customization-with-splits - Subject: Replit Agent Replit introduced Splits, a feature that allows users to fully customize the layout of the Replit Workspace by dragging and dropping tabs or panes to split, merge, maximize, or float them. The update aims to support different working styles while keeping the default layout simple for beginners. ### OpenAI — DALL·E now available without waitlist - Company: OpenAI (openai.com) - Announced: 2022-09-28T07:00:00+00:00 - Category: availability-change - Coverage: not counted - Announcement: yes - Group: announcements - Source: https://openai.com/index/dall-e-now-available-without-waitlist - Record: https://forck.live/items/891-dall-e-now-available-without-waitlist - Subject: GPT / ChatGPT / API DALL·E is now available without a waitlist, allowing new users to start creating immediately. OpenAI cites lessons from deployment and safety system improvements as enabling wider availability. ### Hugging Face — Image Classification with AutoTrain - Company: Hugging Face (huggingface.co) - Announced: 2022-09-28 - Category: capability-change - Coverage: not counted - Announcement: yes - Group: announcements - Source: https://huggingface.co/blog/autotrain-image-classification - Record: https://forck.live/items/2149-image-classification-with-autotrain - Subject: Platform Hugging Face announced that AutoTrain now supports image classification, allowing users to train image classification models with no code. ### Hugging Face — How 🤗 Accelerate runs very large models thanks to PyTorch - Company: Hugging Face (huggingface.co) - Announced: 2022-09-27 - Category: not stated - Coverage: not counted - Announcement: no - Group: routine - Source: https://huggingface.co/blog/accelerate-large-models - Record: https://forck.live/items/2150-how-accelerate-runs-very-large-models-thanks-to-pytorch - Subject: Platform Hugging Face's Accelerate library leverages PyTorch's meta device to load and run very large models that do not fit in memory, using techniques like empty model creation and device mapping. ### Hugging Face — SetFit: Efficient Few-Shot Learning Without Prompts - Company: Hugging Face (huggingface.co) - Announced: 2022-09-26 - Category: research-paper - Coverage: not counted - Announcement: yes - Group: announcements - Source: https://huggingface.co/blog/setfit - Record: https://forck.live/items/2151-setfit-efficient-few-shot-learning-without-prompts - Subject: Platform Hugging Face, with Intel Labs and UKP Lab, introduces SetFit, a framework for few-shot fine-tuning of Sentence Transformers that does not require prompts, achieves competitive accuracy with small labeled datasets, supports multilingual text classification, and trains significantly faster and cheaper than comparable methods.