TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Show HN: JuliusJS – Speech recognition in JavaScript

202 pointsby zzmpover 10 years ago

16 comments

zzmpover 10 years ago
There is now a (very rudimentary) demo on the GitHub page: zzmp.github.io&#x2F;juliusjs<p>Much thanks to @iffy for writing the first pass.<p>It uses voxforge&#x27;s sample vocabulary, so you&#x27;ll need to say things like &quot;Dial 1 2 3&quot; or &quot;Call Kenneth McDougall&quot; for it to understand you, but the vocabulary is easily swapped out for your own projects, as explained in the README.
ar7hurover 10 years ago
Thanks for sharing, nice work!<p>Quick question the Julius website says there is no English acoustic model available [1], how did you solve this? Do you provide a default acoustic model?<p>[1] <a href="http://julius.sourceforge.jp/en_index.php?q=en_grammar.html" rel="nofollow">http:&#x2F;&#x2F;julius.sourceforge.jp&#x2F;en_index.php?q=en_grammar.html</a>
评论 #8406060 未加载
zzmpover 10 years ago
Creator here - I ported this over from the open-source Julius using emscripten. AMA
评论 #8405806 未加载
评论 #8407528 未加载
评论 #8405533 未加载
dansoover 10 years ago
This is sweet...to get an idea of how much fun this could be for web apps, check out the Annyang library (<a href="https://www.talater.com/annyang/" rel="nofollow">https:&#x2F;&#x2F;www.talater.com&#x2F;annyang&#x2F;</a>), which wraps around the Google Web voice recognition API...it works very well, but of course, is subject to Google&#x27;s terms...so an open source system is very welcome
评论 #8407125 未加载
yeukhonover 10 years ago
Pretty cool! When I did my project I had to use <a href="https://github.com/kn/speak.js" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;kn&#x2F;speak.js</a> which is an amazing library. The library still works on Firefox 30, 31 by the time I finished my project (and the project itself hasn&#x27;t change much for a year or two!).<p>I would definitely give this JuliusJS library a try. I am actually amazed that JuliusJS doesn&#x27;t carry all the heavy data like speak.js does (multiple languages support though). I love the fact that you state 100% client side!
hugozapover 10 years ago
Great work, this will be another cool library i star on GitHub and never do anything about it :&#x2F;
评论 #8406906 未加载
bubeeover 10 years ago
Nice work. Can it return confidence scores? Say I want to load 3 commands in my page: 1. Click blue button 2. Scroll down in the yellow text area 3. Expand image of man I feed those to the engine, and when somebody speaks, I get a confidence score on each word so I can determine with a level of configurable certainty that the user is using the command: {click: 0.9878 confidence, blue: 0.8789 confidence, button: 0.1889 confidence)<p>Something like that...
评论 #8406718 未加载
jergasonover 10 years ago
I&#x27;ve played with pocketsphinx.js a fair amount, but this looks WAAAAAAAAAY easier to set up and consume. Nice work.
评论 #8405636 未加载
Gonzihover 10 years ago
Is there online demo anywhere on the web?
评论 #8405758 未加载
sunsuover 10 years ago
Can you use any of the CMUSphinx compatible language models with this, or is there a tool to convert them to something Julius supports?
评论 #8409794 未加载
bikamonkiover 10 years ago
Can this be used to detect a voice&#x27;s unique digital signature? For example I just say my name to login into a website?
评论 #8406406 未加载
cue232sover 10 years ago
Does your application need a nodejs backend for this library to work?
评论 #8406878 未加载
borplkover 10 years ago
Genuinely not sure if the demo is a joke or not.<p>I said &quot;hello&quot; it said &quot;DIAL OH OH&quot;.<p>I said &quot;Apple&quot; it said &quot;GET KENT&quot;.<p>WTF?
评论 #8409801 未加载
cssandjsover 10 years ago
This is awesome - who cares about Windows 10 or Linux, Javascript is the new OS.
评论 #8410179 未加载
CmonDevover 10 years ago
Wow, speech recognition in a Turing-complete language. Amazing.
kelvin0over 10 years ago
This is the kind of technological challenge which must be fun to complete. And it must be quite satisfying for the author. However, whenever I see a &#x27;XYZ in pure javascript&#x27;, I keep getting the impression we are only delaying the inevitable moment browsers have to step to a superior language. Kinda like instead of quickly ripping off a bandaid is better than slooowwwwllly removing it ....
评论 #8406021 未加载