- Vocapra - di.unimi.it

Iscrizione per Venerdì 13 Gennaio
by Claudio Mariani 06 Jan '23

06 Jan '23

Buongiorno con la presente vi chiedo la possibilità di iscrivermi all'evento "go vocapra" del prossimo Venerdì 13 Gennaio. Non sono un allevatore, ma un importatore di genetica (seme ed animali) e quindi non so se sono autorizzato a partecipare all'evento. I miei contatti sono Claudio Mariani Genesi Project SrL https://www.goatimprovement.com/it/ +39 335 5321140 Cordialità C. Inviato da Outlook per Android<https://aka.ms/AAb9ysg>

2 1

Numero di istanze con etichette
by Luca Andrea Ludovico 20 Jan '22

20 Jan '22

Ciao a tutti. Ho creato una nuova pagina nella parte di backoffice, chiamata “Numero di istanze per tipologia di evento” e raggiungibile dalla home dell’area riservata. Per ora, si tratta di un mero conteggio basato sulle etichette. Quando avremo una gerarchia ad albero, sarà possibile anche organizzare i conteggi per categoria, andando nella direzione menzionata stamattina. Devo fare una precisazione: in realtà, non si tratta del numero di audio corrispondenti a una certa etichetta, ma del numero di eventi inseriti con tale etichetta. Per intenderci, se viene segnato un evento di stalla “distribuzione di foraggio” dalle 12 alle 12.30, questo conta +1 nella mia pagina, ma gli audio collegati potrebbero essere da 0 a n. Inutile specificare che la query per estrapolare il numero di audio connessi sarebbe molto più complessa e onerosa. Però, discutendone con i colleghi, questo modo un po’ naif di fare i conti potrebbe andare nella direzione voluta, ossia differenziare al massimo le registrazioni corrispondenti a una certa etichetta. In altri termini, se vogliamo capire come bela una generica capra ferita, è meglio avere 50 capre ferite diverse (che magari belano ciascuna tante volte) rispetto ad avere un unico evento con associati 50 belati da parte dello stesso soggetto. E comunque, nessuna capra è stata ferita in questo esempio che ho appena inventato :) Un caro saluto. Luca

1 0

sintesi delle ultime due riunioni
by Silvana Mattiello 17 Dec '21

17 Dec '21

Cari tutti, vi invio in allegato la sintesi delle ultime due riunioni, in cui sono stati discusse anche alcune modifiche all’architettura iniziale del progetto. Spero siano utili per aggiornare chi non c’era sullo stato di avanzamento dei lavori. Chiedo inoltre a chi era presente di segnalare eventuali inesattezze o omissioni. Colgo l’occasione per inviare a tutti i miei migliori auguri di Buon Natale e Felice Anno Nuovo! Silvana Prof.ssa Silvana Mattiello Dipartimento di Scienze Agrarie e Ambientali - Produzione, Territorio, Agroenergia Università degli Studi di Milano Via Celoria, 2 20133 Milano Tel. 02 5031 6513 <https://sites.unimi.it/democapra/> https://sites.unimi.it/democapra/ http://vocapra.lim.di.unimi.it/ -- Questa email è stata esaminata alla ricerca di virus da AVG. http://www.avg.com

1 0

meeting
by Stavros Ntalampiras 10 Dec '21

10 Dec '21

Dear all, following the doodle result, the meeting is set for 15/12 at 10:00. We'll do it at the usual room of our department. thank you, have a nice day, Stavros -- *********************************** Dr.-Eng. Stavros Ntalampiras Associate Professor, Department of Computer science University of Milan, Italy https://sites.google.com/site/stavrosntalampiras/home ***********************************

2 1

Vocapra Architecture Review
by Mael Vittorio Vena 03 Dec '21

03 Dec '21

Hi everyone, I spoke to my neighbor (the electronics engineer). What emerged is that STmicroelectronics no longer supports WIFI among its wireless technologies. For this reason, we computer scientists met and thought about a reinterpretation of the project architecture, which greatly simplifies all development. The idea is to use Raspberry PI(https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&cad=rja&u… ): a microcomputer to be left in the farms that takes care of doing the classification directly on the farms! In this way there would be no more need to send audio files over the internet (extremely expensive operation in terms of bandwidth), but only the labels that are recognized. To simplify this, I have created two diagrams that show the past and the future of the proposal. I am also attaching an article(https://pub.dega-akustik.de/ICA2019/data/articles/000660.pdf ) that shows how the use of this type of solutions with microcomputer is feasible and already used. let me know if the new architecture we propose is clear to you kind regards, Mael

2 1

Re: [VOCAPRA] R: Meeting about the annotation process and future steps
by Stefania Celozzi 23 Nov '21

23 Nov '21

Dear, here attached the Gantt chart that we have proposed in view of the remodeling of the project. I remind you that we had discussed it in a previous meeting and unanimously approved. See you tomorrow. Stefania P.S.I have copied Tyfenn because she is not on the mailing list Il 23/11/21 10:53, Stavros Ntalampiras <stavros.ntalampiras(a)unimi.it> ha scritto: > > Dear all, > > Here is a list of the topics to discuss tomorrow: > > > App > 1) farmers interaction => can farmers enter information to the app or can they only receive it? > 2) usability of the app => in the design Mael sent, do you think it is usable enough? are there too many or too few features? > 3) Design => opinions on design? > 4) do we need to add/remove something? > 5) deadline to have the app ready? March? > > Annotation > 1) how many files can you annotate per week? > 2) the frequency of each day of the excel? => it would be better to avoid having 100 audio files one day and 0 the day after, so how would you like the file frequencies for each day? > 3) dynamic filter on the number of files per day => instead of setting a static filter (eg goats> 0.9) we could say: give us the best n results for each day > 4) how many classes are there? => considering that we need more than 100 examples per class to train the recognition model > > Project architecture > 1) we need wifi to get the project going, which farms have wifi? > 2) the sensors are not equipped with wifi => since they don't have it, how do we do it? do we buy wifi cards to stick? > > Mael's Contract > 1) is there any news? > 2) PhD + contract is it possible? > > > a domani, > Stavros > > > On 22/11/21 09:31, Silvana Mattiello <silvana.mattiello(a)unimi.it> wrote: > > > > > > > > > > > > > > > > Dear all, > > > > I think that I will be able to attend the first part of the meeting, but then I have another duty and I’ll probably leave early. > > > > However, the presence of Stefania and Tyfenn is much more important than mine for this meeting! > > > > Thank you Stavros for organising the meeting, it is a good idea! > > > > Ciao > > > > SIlvana > > > > > > > > Prof.ssa Silvana Mattiello > > > > Dipartimento di Scienze Agrarie e Ambientali - Produzione, Territorio, Agroenergia > > > > Università degli Studi di Milano > > > > Via Celoria, 2 > > > > 20133 Milano > > > > Tel. 02 5031 6513 > > > > https://sites.google.com/site/stavrosntalampiras/home(https://sites.unimi.i…" rel="noopener noreferrer noopener noreferrer" target="1">https://sites.unimi.it/democapra/ > > > > http://vocapra.lim.di.unimi.it/ > > > > > > > > Da: Vocapra [mailto:vocapra-bounces@di.unimi.it] Per conto di Stefania Celozzi > > Inviato: sabato 20 novembre 2021 13:57 > > A: Stavros Ntalampiras <stavros.ntalampiras(a)unimi.it>; vocapra(a)di.unimi.it > > Oggetto: Re: [VOCAPRA] Meeting about the annotation process and future steps > > > > > > > > Dear all, > > > > > > I will attend the meeting on next Wednesday. > > > > > > Stefania > > > > Il 20/11/21 10:58, Stavros Ntalampiras <stavros.ntalampiras(a)unimi.it> ha scritto: > > > > > > > > > > > Dear all, > > > > > > > > > > > > > > > > > > now that we have a filtering algorithm, we think that it might be good to fix a meeting and understand the current status of the annotation process and see the future steps. > > > > > > > > > > > > > > > > > > I propose to meet at our Department on Wednesday morning at 10:00. > > > > > > > > > I would say that not everyone is required to be present; only the persons who are actually doing the annotation should be there but ofcourse everyone is welcome to join. > > > > > > > > > > > > > > > > > > best regards, > > > > > > > > > Stavros > > > > > > > > > > > > -- > > > > > > *********************************** > > > > > > Dr.-Eng. Stavros Ntalampiras > > > > > > Associate Professor, Department of Computer science > > > > > > University of Milan, Italy > > > > > > > > > > > > *********************************** > > > > > > > > > > > > > > > _______________________________________________ > > > Vocapra mailing list > > > Vocapra(a)di.unimi.it > > > http://mailman.di.unimi.it/mailman/listinfo/vocapra > > > > > > > > > > > > > > > > > > (http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_cam…" rel="noopener noreferrer noopener noreferrer" target="1">Mail priva di virus. www.avg.com(http://www.avg.com/email-signature?utm_medium=email&utm_source=… ) > > > > (#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2 ) > > > > > > > -- > > *********************************** > > Dr.-Eng. Stavros Ntalampiras > > Associate Professor, Department of Computer science > > University of Milan, Italy > > https://sites.google.com/site/stavrosntalampiras/home > > *********************************** > > > _______________________________________________ > Vocapra mailing list > Vocapra(a)di.unimi.it > http://mailman.di.unimi.it/mailman/listinfo/vocapra > >

1 0

Re: [VOCAPRA] Meeting about the annotation process and future steps
by Stefania Celozzi 23 Nov '21

23 Nov '21

Dear all, I will attend the meeting on next Wednesday. Stefania Il 20/11/21 10:58, Stavros Ntalampiras <stavros.ntalampiras(a)unimi.it> ha scritto: > > Dear all, > > now that we have a filtering algorithm, we think that it might be good to fix a meeting and understand the current status of the annotation process and see the future steps. > > I propose to meet at our Department on Wednesday morning at 10:00. > I would say that not everyone is required to be present; only the persons who are actually doing the annotation should be there but ofcourse everyone is welcome to join. > > best regards, > Stavros > > -- > > *********************************** > > Dr.-Eng. Stavros Ntalampiras > > Associate Professor, Department of Computer science > > University of Milan, Italy > > https://sites.google.com/site/stavrosntalampiras/home > > *********************************** > > > _______________________________________________ > Vocapra mailing list > Vocapra(a)di.unimi.it > http://mailman.di.unimi.it/mailman/listinfo/vocapra > >

3 3

Meeting about the annotation process and future steps
by Stavros Ntalampiras 22 Nov '21

22 Nov '21

Dear all, now that we have a filtering algorithm, we think that it might be good to fix a meeting and understand the current status of the annotation process and see the future steps. I propose to meet at our Department on Wednesday morning at 10:00. I would say that not everyone is required to be present; only the persons who are actually doing the annotation should be there but ofcourse everyone is welcome to join. best regards, Stavros -- *********************************** Dr.-Eng. Stavros Ntalampiras Associate Professor, Department of Computer science University of Milan, Italy https://sites.google.com/site/stavrosntalampiras/home ***********************************

2 1

Update on the audio classification algorithm
by Stavros Ntalampiras 29 Oct '21

29 Oct '21

Dear all, We are proceeding with the algorithm which is going to filter the data to be annotated. It works quite well for goats and young goats. The problem arises with the sneezing sounds which are very similar with background noises. Please see some examples to see how alike these are in the following link https://we.tl/t-CG92O2COLl. It is really difficult even for humans to understand the difference between sneezing and background noise. So we need to think how important the sneezing class is, at least for this stage of the project where we are trying to annotate the available data. If you agree, we can discard sneezing for now and continue with the annotation of the rest interesting classes. And then at a later stage, we can try to improve the sneezing detection if you think it is important. Let us know you opinion on this. We may also set a meeting to talk about this. best regards, Stavros Ntalampiras -- *********************************** Dr.-Eng. Stavros Ntalampiras Associate Professor, Department of Computer science University of Milan, Italy https://sites.google.com/site/stavrosntalampiras/home ***********************************

2 1

stavros.ntalampiras@unimi.it ti ha inviato Labeled data VOCAPRA.rar con WeTransfer
by WeTransfer 13 Oct '21

13 Oct '21

stavros.ntalampiras(a)unimi.it ti ha inviato Labeled data VOCAPRA.rar 1 elemento, 200 MB in totale ・ Scade il 20 ottobre 2021 Link per il download: https://wetransfer.com/downloads/d0251332bd123bdd5550a193a3cfd5202021101313… 1 elemento Labeled data VOCAPRA.rar - 200 MB [Ottieni di più da WeTransfer, scegli Pro](https://wetransfer.com/pro?trk=WT201704_email&utm_campaign=WT_email_tr… Informazioni su WeTransfer: https://wetransfer.com/about Aiuto: https://wetransfer.zendesk.com/hc/en-us Informazioni legali: https://wetransfer.com/legal/terms Per essere sicuro di ricevere le nostre e-mail, aggiungi noreply(a)wetransfer.com a [i tuoi contatti](https://wetransfer.zendesk.com/hc/en-us/articles/204909429).

1 0