 representative professionally I'm a front-end developer so today I'm going to speak about common wise it's a Mozilla project so here is our agenda that we are going to speak in the next few minutes so first we will go with the Mozilla and its mission on common wise and then we go with the view of common wise and we will discuss the part of contributing with this project. So Mozilla, what do you know about Mozilla? Apart from the browser. Anyone? They had 5 Fox OS. Mozilla is a non-profit organization that develops Mozilla products that are promoting free open software and open standards. We believe this act of human collaboration across an open platform is essential for individual growth and collective future. So Mozilla's products includes Piper web browser, common wise, 5 Fox OS and deep pitch and we are also going with IRL, Rust, Rust development, VR and AR development and Audance. There are things across and Thunderbird, Emac line. We have many multiple products in the Mozilla. So let's discuss about the common wise. So after deep pitch, Mozilla has been come up with common wise project. It's a data model where we can use this for open resource for our project development in our wise development. So basically when we need to work up on the wise recognition do we think it's whether we can accept it's accessible to everyone wise? Speech data models? It's not to get it. Even it is available. It's cost about like ten to thousands of dollars. So it's not easily available to everyone. There are many data models like library speech and corpus. But the thing is like it's like a proprietary solutions. Everyone is up to the property and we do not have any privileges to accept everything we need to. We do not get that. So Mozilla has been come up with the introduction of common wise that we need to be available for everyone. And we are the one who wants to get contributed and we are the one who wants to develop that. So that's a thing. Here is the main agenda that we are going to speak about that. So this project has been launching in the year 2017. So right now we are going with the only one and English version. And in future we are going with the multi language support. So main agenda is like we want like users in the visitors can use what contributors can do. So Mozilla is like the first thing is like regarding the perspective of technical or non-technical. We can donate our voice. There are multiple sentences on the website where you can donate your speech. And the next thing is like you can provide your sentences which can help us any other person to get recorded. Like there are movies, speech, conversations, and general talks, everything that you can provide. So we want developers to build amazing things. And we have a we can like we can make like from voice administration to access from all over the world. So what the first thing is like I have discussed in the earlier thing is that you can donate your voice. You're able to say like the sentence like ladies and gentlemen, the wagon is going to be the detach. So this is one of the sentence that we can able to donate our voice. And the first thing that first thing is like you need to create your profile in that like your accent and your state of state of language and accent like US and Singaporean English, Malaysian English. So the reason is like we need to filter about the data speech models built upon your language and accent. So the reason we have been introduced this profile. And so once you get once you get recorded ladies and gentlemen, this wagon is going to be detached. So for the first three multiple decision. So you will you will be able to review the review your data, review your recording, and you can able to submit that and it will be stored according to your accent. The other way is like you can, you can provide your sentences like movie scripts, conversation, speech, and any of the sentence like is that here is a criteria like you can write yourself and not a public based thing from somewhere else. And you can also make the sentence like conversation as easy to read. And there's a license like CCO CCO license that you need first before that you need to review the data partner and then you can submit that and 50 sentences per link. And generally you can use an open source model like the database and you can provide your link in the link in the video is just like a few of the samples where sentences have been encoded into this. The other thing that you can do that is like you can validate your sentences, validate other sentences. Generally, the person who have been recording the voice, there may be some sentences, but there may be problem with the recording voice due to some noisy and some cosy places. So there will be a problem with the accent or reading the sentences in a proper way. So we have a record, we can validate the recorded recorded sentences and the snippets will be validated based upon our validations, multiple validations. So it will be approved once everything is done. So here is our data set like this has been data with 200. We have right now in our data model, we have four, four like recordings with 200, 200 people with the final speech. So this is this is one of the largely available data set which is available for free in the market, which is which way we are providing for people who wants to contribute for a speech recognition system. This is one of the one you can be able to use and you can you can you do not have any you can use it as per your property or for your solutions. And the other thing like if you if you want to if you want to get contributed for through technically by coding, this is the one that you get requirements like you need to have a node.js knowledge and npm with version four and and yeah, with a big version one higher. And I think it's like like you need to you know, we can also use the FF MPG for developing the library speech like multimedia speech library and my is pure or my my idea for using storage of voice to storage. Generally, the common voice is having about Amazon S3, which we are showing for the clip storage that you can get you can also get contributed. You can also get configured using the GitHub. I will show you that in the next. There is the one like you can you can also configure for storage of clips into your local. So you need to first log first need to create a account in S3 Amazon S3 and you need to create the bucket and choose a bucket and depending upon a region you can go and and you need to set the properties. Everything is here in the cleared up clearly documented and you have a configuration there is where you need to run your voice into run your local into your local that local mission. Like you can also submit your bugs if you can if you can encounter encounter any using that like it may be about operating system on it may be right now this voice recognition can be done using the online web and we have online iOS app. So we are working on the Android application where expected to release in the soon in the next coming quarter. This is a start from from starting from July 2017 from the project where project has been started and from the till today to today. This is the one hour starts from 200 we have a 200k total visitors and 450k recording so far now and 890k business. So and 600 hours of spoken words. You can also you can also connect to all the developers and you can get interacted with them and you can if you have any query you can also get connected with them using the Slack channel and also you have a discourse topic where everything related to common wise and other process will be placed here. Everything get updated and where you can get up to keep updated with this stuff related with the current with this current project but from the common wise project you can also involve like if you have been put in JavaScript you can also involve web development and add on development and here is a website like like you can go with that and what I what can I do for Mozilla.org based upon your your skills you can you will be getting a master's skills and your how to proceed further everything is clearly presented in there and these are the areas of contribution where you can get involved to the Mozilla like if you are being very well versed with the support usage like like helping others you can you can get into the support Mozilla project and if you are being very well versed with the bugs and testing and you can involve with the bug fixing testing and bug tracks there are multiple tracks for every week where you can get involved with that and you can involve and marketing we have a marketing team where you can get involved with that and as an affiliate and if you are being in a localization you want to get contributed to your to change your talk to change your talk there are multiple articles where you can knowledge that where you can change your documents into your local language and web development like the UI and UX which you can help in developing a common wise project website and marketplace like you have a your family with Chrome like Chrome apps like the same we have in the Mozilla add-ons we call it using a using knowledge of having an knowledge of JavaScript is fine to work with the add-ons you can you can develop your add-ons with an easy software and we are you can involve in a VR and trust development. Any questions? I have a question so so far how many of the English words available are you able to translate like are you able to like get the data form because they're like there are like a sort of like infinite number of words that people can speak so far how much like of the like whole language are you able to cover also is this just for English or is it for right now this is for this project is for the only English we are working with the multiple languages so far. Okay. So we are expected to going with the last year last quarter. So like the data model is up 200 k words 800 sorry 400 k words for the English model. Okay. And with that 200 people of listeners who wants to be a contributor for we are only we have a project for iOS and web. So far like how much of like the vocabulary are you able to complete like you said that there are around 400 k words right. Yeah. So how much like how much of that are you like here do you have the data for right now. Yeah. Model which I have been told you like with the 400 k data model. Okay. Where you have a speech of around like one GP and it means like I cannot I should I cannot give you the clear one like a number it may be around almost five to six lines of words. Okay. With those recordings for the only English. Okay. So this is like available for everyone to download and you want to download and but you need to use before that you can go with the zero license. Thank you. Thank you. Any other questions for Sandy. Thank you so much.