 Če je bilo v svojom. Zelo si počke, da smo vse zelo zelo. Zelo, da smo v svojom komunitih zelo, tako da so počke, da smo nekaj neči sveti. Oče so pričeli, da smo vse zelo, da smo vse vzeli, da je zelo, da smo vse zelo. Mozila komunitava je ne moj najbolj, je ne moj najbolj, je najbolj, kaj je v tem, da je skupaj zelo, bo se je objevajalo, da se je koncentruje in češtje kreda, da se je koncentruje in češtje kreda, kaj je je, je, je, je, je, je, je, je. Zelo vse veče je vse kreda, da se je dobro kreda, da je dobro kreda. kot tudi nalega. To istajno je odوse med vse dolačit. Spostakam o nekaj bones. Z vsem, da je vse projekta u gdje Mozzille Italia komunite. Pati če je Mozzille Italia komunite? Menejno tudi mozzille vse u tudi studento co je istati tudi. Sve nedaj so prišta. Opočijo je online, ko v načaj igram razdavljati o linguistijce, in da vsi se evočni odličenih povodov, da se je zelo vsi in vsi in vsi in vsi in vsi splivni odličenih povodov na veči odličenih povodov. Vsi je in vsi, vsi, vsi, vsi, vsi, vsi, vsi, vsi, vsi, vsi, vsi, vsi. Prvo se ga učem obladi v teresnih ste in vsi. Tukaj v sej seara je tukaj izgledan, tukaj tukaj izgledan vsi, tukaj veči nekaj izgledan. Tukaj vsi je zelo vse. In, da se pošličimo, zelo smo v plenih dve roli, ker spremljenje, mozila, je spremljenje, v pošličnih dve roli. Pistofol, te data. To mozila se vzela v komandu projekti, vzela.mozila.torki projekti, kjer komunitičnih, pošličnih pošličnih pošličnih pošličnih pošličnih pošličnih pošličnih pošličnih pošličnih. Prevali sem. Si bo whasila v tom uko superalivni set, ta kjer je spl wrist']. Zelo smo si cinema madmu, ni š sergeova, ni ustojvajtez, ne poplu tudi nekako se zelo ste p slotimo. Nà začali straž ... Jel jim na najskega dve rotinga nam teúpili, vse je všeč vzivno, da si se niče zaprejče na vse, dobro se regovorjamo mnogočne, možno ozila razljadne, zato se odpravila, nekaj je projeg tega, zato se neče začal. V prvi delu izvajemo to, začal tudi projekti, počutne dato. V sevih delu, zelo smo zelo začal, načal je, skupaj. Zato, nekaj se poznali, načal je, začal je, začal je, načal je, načal je, načal je, jedin moj that we guarded a lot of sentences and defined some rules. As an example, how much we need to belong because we discovered when we tested these because we gathered like eleven thousand sentences and we saw when we did the first events about recording that we had some problems that like long sentences because Italian has long words so we didn't test them ...zanoj, da ne zaredamo sreč, da se bo krati... ...pravih... ...sega, in tudi, da ta očoče v vladi moramo... ...daj, da smo pošli drugi pošli, da smo pošli... ...ač sem, da bo vladi... ...zanoj, ki bi smo pošli... ...zanoj, da bi smo pošli... ...ko 20 vradi... ...pošli si možno, da bi smo pošli... ...zanoj, da smo pošli... ...ko so prišli, da smo pošli... ...nača vzupi... ...tak je vše... ...zanoj, da smo pošli... Haji, naš don, nami ne so nerega reš коji me jede Gwen LO. Me the next step was to experimenting promotion with zero funds because we are a community, we are volunteering, we do something else in graduating. We have so many money put on that, so we experimented different ways, like organizing events in different cities with different people as speakers so we can have something in the ground iz nekaj počudov, da zelo smo v initalijskih slijdetku z 40 slijdetku, z mnogožem vse. Zelo, da svoje smjeda je, da te slijete so vživite, da se nekaj nekaj, nekaj sem sem si vse vse tudi prav na universitetu in v computeria, ki so vsem počudov v štam zelo. Proto, da je inočen začar, nekaj nekaj nekaj nekaj da se nekaj nekaj ne bo noga o taj, so nekaj biti nekaj semči čutke. Tko v rukšenju modularu, jih sem zelo všem soli seženje tudi, da je vse zelo, da ta komunita izvijaja za njihovo, še we začajali. Tote tudi je tudi čutke jamške. To je vseženje, tudi je vseženje vseženje, je začajal, nekaj vseženje, nekaj vseženje, tudi nekaj vseženje. To je se svoj, ki sem izpronil, nekaj je vseženje, nekaj je vseženje, ki se glede, da je tudi poštite, da je od vrstej do vrsta. Na vrst, da smo se poštite, kako je vrstej do vrstej, zelo je to je vrstej. V našem sej, kaj smo pačili? Srečak smo poštila vrstvenje in provošnjene, ker smo droge oga tudi zelo. V Italijanii sem vsozil in 60 milijos priče, ker je to v zelo. zelo smo testi vnev socijalne taktike z vseče. The first of all was to write a big kind of weekly post on the forum with a lot of updates from international and national about common voice and speech. This is because we got always new people joining on our channels asking always the same questions. So we said maybe we need to save time to reply to these people doing something else because we have too many things to do. And this work at very well because during the time we appreciated a lot when we passed the new year how many things we did. And we shared this post in a lot of new place that we never tried during the years like Reddit in the Italian subreddit about ity world and also specific Facebook Italian groups where we wasn't, there were, there are taking into sastics. So not just developers open source people, we need to find new people just to get the most amount of feedback that we can. And also with the help of Mozilla we got the promotion inside about home page of Firefox. There are the snippets. I don't know how many of you use so, but there was a specific snippet for Italian user about try common voices. For us was kind of incredible because we did it in common voice in five months, 10 hours, 10 hours of recordings. With two weeks, with this promotion we got 40 hours. Just to say that for us was eh, it's kind of incredible what we did with this kind of a bit of promotion. Very good. Also we created a category in the discourse of Mozilla that is public, is international. You can find also Mozilla employer, you can find all the community and there is a common voice category and also this speech. And we got an Italian category where to speak Italian just for us because the common voice website referred to this course so there was the only link where people want to try to find us. And also we tested with specific kind of tweets but not from our Mozilla Italia social channel official but also ours with different kind of keywords, hashtag just to see what was working better. One of the things that we did was to keep updated as live because time comes, there are changes, project evolving so we need to update it. And we got new people that was using it and we have no idea that contributed back to improving it. Next also we created specific kind of activities for people with zero knowledge of Mozilla, what works, what is the project and also with few time to contribute. We saw looking at the people that was joining us, what was the most interested things for them, what was the most uncommon background, these kind of things. Just understand what the people was joining, what we can offer just to them to keep them part of what we are doing. The second part was to thinking about documentation and planning. So mainly community management. We define a draft in English, this time not Italian and this public of course like the rest of staff about our needs from Mozilla and the project Common Voices mainly. And we started to update this document during the year with all of what we did, what we cannot do, what are the problems, etc. And the same time we shared that because it was public outside the community with these new channels that we found. Again to be sure that everyone was informed but that's to get the most feedback to understand or to explain what we are doing because it's kind of new or everything. And we planned a meeting with Mozilla employees that follow Common Voice just to share these our needs from the community and people outside that have no idea what is Common Voice and to get answers and fix some things. So for us it was very important to have this one-on-one relation with Mozilla and this is possible. Also review and gather ideas from all over the communities not maybe just me or another volunteer that say okay we have to do this. No, we was open to everyone just to have the best effect that we can do with what we have a tiny community. And we do what's usually a monthly video call that is open to everyone and we publish a YouTube after the call but it's open to everyone so we got people that was just joining there just to be to show what we are doing just you know people and define goals for any kind of area that want to do something. Later when we saw our people our volunteers what they can do we saw the priority so just not say we are these are the area but just to say for every area we want this as a priority and let us see who can do it everything. So for us was very important also that every decisions was open so everyone can say something because for Mozilla we need transparent we need inclusion of everyone and also we extended our internal ideas to all the community because we open the thread in the international category on this course to see also other community what how was doing something that maybe we have an idea maybe someone already did and often happens but there is no communication often. The third part was maybe the most one that we are interested as developer like me is development and experimentation. First of all we created a repo that's based on the scripts from the French community of Mozilla because they have a Mozilla employee that is Alexander alexandre jo Alexander that is watching me right now probably and that is working on this pitch and he written a lot of bash script and docker image to generate the model for French so we said well we can fork it change what we need just to use Italian stuff like files corpus this kind of things and this was by just one person me that doesn't have so many experience about how works this pitch I am a developer I work bash this kind of things but they have no idea how works machine learning in the details so I started just because it was time the cool part that when I did everything fix the batch the batch script to do for Italian I said to the whole community to all these new social groups this kind of things we need testers when I say that a lot of people joined on testing it on their company web server of West Azure that are very accost so they just to try everything and contributed back to fix what wasn't working what was can be improved excited this kind of things in that way we got new valid maintainers more than me skilled of this kind of things they are now are pushing on this kind of things also for me was very important but also intriguing and thrilling because after seven years in San Mozilla community we got for the first time companies that was contributing to us to our project in this case we got patch server support experimentation in this case the most simple is was a telegram bot that compare the recording with google speech and our model for us was really now we can play with something without a style of speech because it's not easy and now me because the first person is just doing project management and there's something about the development of this without following that part because I want to say that because is not so scary getting a model from who doesn't have knowledge you just need to start and ask because the community specific of a language it's big and you can find more easily later talking with our community we built an android app for common voice because there wasn't because it's our website because for the kind of reason also to spread at must we can we did a new left flat because we have this repo about in italian with the left flat about I don't know Mozilla what is Firefox this kind of thing so when there is an event by I don't know a linux user group in a city and we don't have anyone there they can printing a chart and we did one about common voice at this pitch and during these years we contributed a lot in different common voice tools and project buzzed on our needs on problems that we find so this was the first big achievement for us a fizz public official this pitch model online built in october with the docker image to generate the model so you can do it buzzed in two different data set there are not so many hours and this is a problem that we are addressing but what we need to do now after to have the model well, we have the model but we don't have any instruction or to use it so we need to do a bit of documentation also a week ago there was a new release of common voice data so our data set now is 90 hours and it's still growing so we need to update the model and also with a lot of things like transfer learning this kind of things and testing it a cool things was this because I joined the KDA academy in Milan just because I'm a KDA user sometimes contributor so I made the shift editor of this magazine printed and he interviewed me about common voice of what we was doing in the first time after Firefox first moment we got something in a magazine talking about what the community was doing it there is also a video interview uploaded on YouTube but that's in the DVD attached to the magazine so for us was very interesting because we did any effort on doing this kind of marketing just going around for us was very helpful because I have no idea that I got this for the community so what is next this year for itali will be very important because we will have two big international national international events and the past year there was nothing just to understand what is the situation and we will join them with different ideas about testing it improving etc we will want to create also italian videos about how to create to use common voice because we have people that always ask the same questions again like what is the best microphone I have to do a review a sentence as an example so after a while is better to do a video that is more easy to see and instead of long documentation also we need to do the tutorials about using our model and we want to do it in italian but there are a lot of projects that already implement this pitch and release these tutorials but we want to do it in italian for our model that way we can gather more people and also get you ready for the linux day because in italy we don't have the software freedom day we have the linux day it is usually in 70 cities in the same day and as Mozilla we have usually five or six cities with Mozilla talks we want to be sure to extend this our area so closing it there is a long block post version with all of the links also in the forum because the Mozilla employees that's working on come on boys ask at me to write my experience and edit it and it also the slides so you can find a lot of links ideas etc. out there for any question of course I'm here instead if you are italian want to know more we have telegram groups you can reach us with these both in case for the others non italian speakers of course them here and finish it do people have questions no questions good okay to go ah okay one this is a cool so the question is how do you how do you manage variation with languages and the example was used as french where when there is accent between regions and they don't use the same expressions and stuff this is a good example because if you see the french scripts they are using different data sets like also for african but the good part is that common voice as when you are register a profile you can set what is your accent for specific languages like french and english mainly for italian is complicated so in the data set is reported for every recording is anonymous the user but you know what is the accent so if you want to build as an example just use common voice for french but from people from poor french as we can say you can do it and what is the quality when you mix different data set in this like this case well this depends on the amount of hours more amounts of hours you put there might be the quality is better so it's kind of experimenting this a part of the answer is you need diversity in the recordings if you listen if you participate a bit and go to voice.mozilla.org and you don't want to use your voice you can just listen you'll hear that it's mainly male that are my age talking there's little variation there so if you have a girlfriend that wants to participate that helps already if you have a girlfriend that's from the south with the proper south accent that helps already et cetera et cetera et cetera the more diverse training data the model has the more it's gonna be able to extend and understand every accent and not usually not talk these accents but at least understand them so it's about the amount of it's about the volume of samples the more the merrier exactly one of the experiment that we want to do is for transfer learning use the spanish model that a volunteer built and see how much change the quality of the tallya one is not the same language but there are similarities and we are now in a phase of experimentation so with transfer learning it's possible to test these kind of things and of course again the big amount of hours you can put on the speech is better the points that you need to timing everything just to test the quality of the tallya model are now is kind of horrible because we did it just in time for the linux there at the time but we need more people that can help us but you can try for french and see what happens I don't speak french so I don't know the quality of the tallya one so is for any kind we need more testing and experimenting because now I think we are at the first steps to see what will happen the good part that is everything open source or public domain so you can use it for everything you want other questions thank you very much grazie