Whisper
readability
Whisper | readability | |
---|---|---|
33 | 52 | |
7,330 | 8,181 | |
- | 4.6% | |
6.5 | 6.3 | |
7 months ago | 1 day ago | |
C++ | JavaScript | |
Mozilla Public License 2.0 | GNU General Public License v3.0 or later |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
Whisper
-
Nvidia Speech and Translation AI Models Set Records for Speed and Accuracy
I've been using WhisperDesktop ( https://github.com/Const-me/Whisper ) with great success on a 3090 for fast & accurate transcription of often poor quality euro-english hours long multispeaker audio files. If there's an easy way to compare I'm certainly going to give this a try.
-
AMD's CDNA 3 Compute Architecture
Why would you want OpenCL? Pretty sure D3D11 compute shaders gonna be adequate for a Torch backend, and they even work on Linux with Wine: https://github.com/Const-me/Whisper/issues/42 Native Vulkan compute shaders would be even better.
Why would you want unified address space? At least in my experience, it’s often too slow to be useful. DMA transfers (CopyResource in D3D11, copy command queue in D3D12, transfer queue in VK) are implemented by dedicated hardware inside GPUs, and are way more efficient.
-
Amazon Bedrock Is Now Generally Available
https://github.com/ggerganov/whisper.cpp
https://github.com/Const-me/Whisper
I had fun with both of these. They will both do realtime transcription. Bit you will have to download the training data sets…
-
Why Nvidia Keeps Winning: The Rise of an AI Giant
Gamers don’t care about FP64 performance, and it seems nVidia is using that for market segmentation. The FP64 performance for RTX 4090 is 1.142 TFlops, for RTX 3090 Ti 0.524 TFlops. AMD doesn’t do that, FP64 performance is consistently better there, and have been this way for quite a few years. For example, the figure for 3090 Ti (a $2000 card from 2022) is similar to Radeon RX Vega 56, a $400 card from 2017 which can do 0.518 TFlops.
And another thing: nVidia forbids usage of GeForce cards in data centers, while AMD allows that. I don’t know how specifically they define datacenter, whether it’s enforceable, or whether it’s tested in courts of various jurisdictions. I just don’t want to find out answers to these questions at the legal expenses of my employer. I believe they would prefer to not cut corners like that.
I think nVidia only beats AMD due to the ecosystem: for GPGPU that’s CUDA (and especially the included first-party libraries like BLAS, FFT, DNN and others), also due to the support in popular libraries like TensorFlow. However, it’s not that hard to ignore the ecosystem, and instead write some compute shaders in HLSL. Here’s a non-trivial open-source project unrelated to CAE, where I managed to do just that with decent results: https://github.com/Const-me/Whisper That software even works on Linux, probably due to Valve’s work on DXVK 2.0 (a compatibility layer which implements D3D11 on top of Vulkan).
-
Ask HN: What is your recommended speech to text/audio transcription tool?
Currently, I use a GUI for Whisper AI (https://github.com/Const-me/Whisper) to upload MP3s of interviews to get text transcripts. However, I'm hoping to find another tool that would recognize and split out the text per speaker.
Does such a thing exist?
- Da audio a testo, consigli?
-
Ask HN: Any recommendations for cheap, high-quality transcription software
I just used Whisper over the weekend to transcribe 5 hours of meeting, worked nicely and it can be run on a single GPU locally. https://github.com/ggerganov/whisper.cpp
There are a few wrappers available with GUI like https://github.com/Const-me/Whisper
- Voice recognition software for German
- Const-me/Whisper: High-performance GPGPU inference of OpenAI's Whisper automatic speech recognition (ASR) model
- I built a massive search engine to find video clips by spoken text
readability
-
2markdown – Transform Websites into Markdown
Why not just use something like https://github.com/mozilla/readability
And not pay $0.01 per request?
There’s a node version too https://www.npmjs.com/package/@mozilla/readability
- Mozilla: Readability.js
-
CSS for readability
I'm working with the Mozilla's readability library https://github.com/mozilla/readability to get the "readable" text from articles and now I want to style the extracted text in a readable way.
-
Building a Serverless Reader View with Lambda and Chrome
Do you remember the Firefox Reader View? It's a feature that removes all unnecessary components like buttons, menus, images, and so on, from a website, focusing on the readable content of the page. The library powering this feature is called Readability.js, which is open source.
-
Webrecorder: Capture interactive websites and replay them at a later time
I wonder if Firefox "reader mode as a utility" might be a viable alternative for Pinboard like "content oriented" archiving?
https://github.com/mozilla/readability
-
Creating an advanced search engine with PostgreSQL
Depending upon the type of content, one might want to look into using the Readability (Browder's reader view) to parse the webpage. It will give you all the useful info without the junk. Then you can put it in the DB as needed.
https://github.com/mozilla/readability
Btw, readability, is also available in few other languages like Kotlin:
https://github.com/dankito/Readability4J
-
Seeking a tool or method to convert webpages into Q&A format using NLP
Use Mozilla's Readability to extract that sweet, sweet text content from webpages.
-
I built a free prompt managing tool - Knit
Same as above but the ability to grab the entire article text (you can use the Readability library for that: https://github.com/mozilla/readability)
-
I need automatic source URLs when I paste any text onto a card or note, like on OneNote.
// Original script // https://gist.github.com/kepano/90c05f162c37cf730abb8ff027987ca3 // Bookmarklet Converter // https://caiorss.github.io/bookmarklet-maker/ // Libraries // https://github.com/mixmark-io/turndown // https://github.com/mozilla/readability javascript: Promise.all([import('https://unpkg.com/[email protected]?module'), import('https://unpkg.com/@tehshrike/[email protected]'), ]).then(async ([{ default: Turndown }, { default: Readability }]) => { /* Optional vault name */ const vault = ""; /* Optional folder name such as "Clippings/" */ const folder = "Clippings/"; /* Optional tags */ const tags = ""; function getSelectionHtml() { var html = ""; if (typeof window.getSelection != "undefined") { var sel = window.getSelection(); if (sel.rangeCount) { var container = document.createElement("div"); for (var i = 0, len = sel.rangeCount; i < len; ++i) { container.appendChild(sel.getRangeAt(i).cloneContents()); } html = container.innerHTML; } } else if (typeof document.selection != "undefined") { if (document.selection.type == "Text") { html = document.selection.createRange().htmlText; } } return html; } const selection = getSelectionHtml(); const { title, byline, content } = new Readability(document.cloneNode(true)).parse(); function getFileName(fileName) { var userAgent = window.navigator.userAgent, platform = window.navigator.platform, windowsPlatforms = ['Win32', 'Win64', 'Windows', 'WinCE']; if (windowsPlatforms.indexOf(platform) !== -1) { fileName = fileName.replace(':', '').replace(/[/\\?%*|"<>]/g, '-'); } else { fileName = fileName.replace(':', '').replace(/\//g, '-').replace(/\\/g, '-'); } return fileName; } const fileName = getFileName(title); if (selection) { var markdownify = selection; } else { var markdownify = content; } if (vault) { var vaultName = '&vault=' + encodeURIComponent(`${vault}`); } else { var vaultName = ''; } const markdownBody = new Turndown({ headingStyle: 'atx', hr: '---', bulletListMarker: '-', codeBlockStyle: 'fenced', emDelimiter: '*', }).turndown(markdownify); var date = new Date(); function convertDate(date) { var yyyy = date.getFullYear().toString(); var mm = (date.getMonth()+1).toString(); var dd = date.getDate().toString(); var mmChars = mm.split(''); var ddChars = dd.split(''); return yyyy + '-' + (mmChars[1]?mm:"0"+mmChars[0]) + '-' + (ddChars[1]?dd:"0"+ddChars[0]); } const today = convertDate(date); // This is the output template // It is similar to an Obsidian core template // except to insert a value we use: ${value} instead of {{value}} const fileContent =`--- type: clipping date_added: ${today} aliases: [] tags: [${tags}] --- author:: ${byline.toString().split('\n')[0].trim()} source:: [${title}](${document.URL}) ${markdownBody} `; // This copies your text to the clipboard navigator.clipboard.writeText(fileContent); // This creates a new document in Obsidian containing your clipping // I commented it out as this isn't what you asked for /* document.location.href = "obsidian://new?" + "file=" + encodeURIComponent(folder + fileName) + "&content=" + encodeURIComponent(fileContent) + vaultName; */ })
- Any js packages to only scrape relevant content from a webpage?
What are some alternatives?
whisper.cpp - Port of OpenAI's Whisper model in C/C++
parser - 📜 Extract meaningful content from the chaos of a web page
whisper - Robust Speech Recognition via Large-Scale Weak Supervision
koreader - An ebook reader application supporting PDF, DjVu, EPUB, FB2 and many more formats, running on Cervantes, Kindle, Kobo, PocketBook and Android devices
TransformerEngine - A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper and Ada GPUs, to provide better performance with lower memory utilization in both training and inference.
hn-search - Hacker News Search
just-an-email - App to share files & texts between your devices without installing anything
readability.php - PHP port of Mozilla's Readability.js
ggml - Tensor library for machine learning
rssguard - Feed reader (and podcast player) which supports RSS/ATOM/JSON and many web-based feed services.
beaker - An experimental peer-to-peer Web browser
SponsorBlock - Skip YouTube video sponsors (browser extension)