Part 1-EDA-Audio Classification Project Using Deep Learning

Поділитися
Вставка
  • Опубліковано 5 жов 2024

КОМЕНТАРІ • 122

  • @krishnaik06
    @krishnaik06  3 роки тому +27

    Make sure you implement till here. Data set will take time to get downloaded

    • @abs412000
      @abs412000 3 роки тому +1

      Now this is really Cool !!! Super Excited for following Videos

    • @abirkhan924
      @abirkhan924 3 роки тому +1

      Put this in Deep learning play list.

    • @hiteshsingh9859
      @hiteshsingh9859 3 роки тому

      sir can you give your telegraph channel ..previous link showing invalid .Thank you

    • @junaidjaved5109
      @junaidjaved5109 3 роки тому

      if meldata file is not available in for datset, what should we do?

    • @lost_soul8711
      @lost_soul8711 2 роки тому

      sir......how to convert our own sound data set to csv file ??

  • @gigsconnect8517
    @gigsconnect8517 3 роки тому +9

    The most clear explanation on AI so far in UA-cam, as I've encountered

  • @xxegyzz5250
    @xxegyzz5250 2 роки тому +6

    thank you so much for uploading this tutorial it really help me a lot. Your explanation is very clear so far i've encountered in yt. Tutorials about audio/sound classification is very rare. I hope that liking your video and subscribing to your channel can help. Please continue uploading videos in the future.

  • @janithdesilva7518
    @janithdesilva7518 2 роки тому +4

    One of the great explanation I ever seen. Could you please do a full video of how we can reduce the noise of a whole audio set ?

  • @mshabanaazmi13
    @mshabanaazmi13 3 роки тому +4

    Thank You Krish.... U r such a great teacher..... U make tough concepts very easy....

  • @artsofdeeplearning902
    @artsofdeeplearning902 3 роки тому +5

    Sir! Very much helpful I got a similar problem statement but I was not able to do it..

  • @souravmohapatra8139
    @souravmohapatra8139 3 роки тому +2

    I got a audio data problem in a recent interview....thanx for this

    • @loltelr6560
      @loltelr6560 3 роки тому

      If u have some kind of educations stuff for ex pdf and books can u send me

  • @rupakdey6753
    @rupakdey6753 3 роки тому +10

    Thank you sir for listening to my request.It means a lot

    • @amit_tiger63
      @amit_tiger63 11 місяців тому

      If I want to create real time project like this then how to create its metadata.

  • @prasadseptember
    @prasadseptember 2 роки тому +1

    I simply love the way you are sharing your knowledge.
    Thank you very much !
    God bless 🙏

  • @Rayiana
    @Rayiana Рік тому

    Honestly, this is the best video that explains. Signal Processing 🤩 Thanks a lot!

  • @fatmademir554
    @fatmademir554 Рік тому

    That's a really instructive, explanatory and beneficial video. Thank you so much.

  • @benbelkacemdrifa-ft1xr
    @benbelkacemdrifa-ft1xr Рік тому +1

    Thanks for this tutorial. Can we do the test using sound sensor?

  • @okopyl
    @okopyl Рік тому

    Could you please explain what is your goal of the project? What is your input for predictions? What is the output form and data?

  • @anuragpatil3820
    @anuragpatil3820 3 роки тому +2

    Much awaited 🙌

  • @2007chandanashish
    @2007chandanashish 3 роки тому +1

    Here we can see the data is almost balanced. But just in case , what could have been done if the data is imbalanced ?

  • @sahilgarg4850
    @sahilgarg4850 3 роки тому +3

    Please try to Upload the remaining parts asap and could you please extend the classification part Abit more by using some more graphs or libraries. That would be helpful.

  • @viewview6687
    @viewview6687 Рік тому

    One of my favorite teachers

  • @arindamroy7671
    @arindamroy7671 3 роки тому +1

    At 10:13 the reason you gave for not getting the error is not correct it seems. You were getting the value error at ipd.Audio(filename) since you did not specify the extension in the filename. It would work fine without the sample rate information that you mentioned is causing the error.

  • @IstiakAhammed
    @IstiakAhammed 2 роки тому

    Thank you so much for making this tutorial for us. It is really helpful for us. I would like to request to you could you please make a video for audio enhancement using deep learning? I will wait for your feedback and expect the video or any suggestions soon. Thanks again.

    • @amit_tiger63
      @amit_tiger63 11 місяців тому

      If I want to create real time project like this then how to create its metadata.

  • @daddyallu1542
    @daddyallu1542 3 роки тому +1

    Thanks a lot sir.Sir, please upload the part-2

  • @adewunmiobajimi7420
    @adewunmiobajimi7420 11 місяців тому

    Thanks a lot... My question is, what is the difference between audio and video mining, and audio ,and video classification?. Or are the two same thing?

  • @dexnug
    @dexnug 3 роки тому +4

    if I make my own dataset not from Urban8k, and how to create the csv metadata?

    • @aditimondal3995
      @aditimondal3995 2 роки тому

      I was thinking the same , have you tried it? I am going to try it.

  • @hadjdaoudmomo9534
    @hadjdaoudmomo9534 2 роки тому

    Wonderful explanation, thank you so much.

  • @shreyasb.s3819
    @shreyasb.s3819 3 роки тому +1

    Very nice tutorial. Thanks

  • @MayoAISpace
    @MayoAISpace Рік тому

    Great video but is it possible for audio data to distinguish persons i.e voice biometrics

  • @shriharimutalik3231
    @shriharimutalik3231 3 роки тому +3

    Sir , are you from gulbarga ..?

  • @mahtabgolshanikia8869
    @mahtabgolshanikia8869 2 роки тому +1

    That was a great explanation. I just wondering what if I have only the Audio files?
    How may I create the CSV file out of that many wav files?

    • @amit_tiger63
      @amit_tiger63 11 місяців тому

      If I want to create real time project like this then how to create its metadata.

  • @rajanikadebnath3404
    @rajanikadebnath3404 2 роки тому +3

    Hello sir, I wanted to ask, how do we extract the number of pauses an audio file contains?

  • @FaizanAli-lw1nl
    @FaizanAli-lw1nl 10 місяців тому

    Great explanation. @krishnaik I want to classify the audio to predict speech/music/silence or background music(noise, applause, etc anything mixed sound) in an audio. how to do it?

  • @mrinalbhardwaj3060
    @mrinalbhardwaj3060 3 роки тому +1

    Thnx sir for uploading this video. 😊

  • @hiteshsingh9859
    @hiteshsingh9859 3 роки тому +1

    can anyone give krish sir's telegraph channel ..previous link showing invalid .Thnks

  • @shivrajak2804
    @shivrajak2804 2 місяці тому

    can i implement a real time emotion detector by refering to this video

  • @mukhlisraza
    @mukhlisraza Рік тому

    Great explanation, really Cool !!!

  • @maddikuntaanilkumar9596
    @maddikuntaanilkumar9596 3 роки тому +1

    where, how can i get real time projects on data science

  • @visakhsikhamani8792
    @visakhsikhamani8792 3 роки тому

    Sir can you make recommendation of songs using the features used for genre classification

  • @manikjain7195
    @manikjain7195 3 роки тому +2

    🔥

  • @harishjk6478
    @harishjk6478 3 роки тому +1

    Wonderful 🔥

  • @pepetisiddhardha9848
    @pepetisiddhardha9848 3 роки тому

    it would have been if some what small size dataset is being used

  • @faresbecheikh7052
    @faresbecheikh7052 Рік тому

    Please how to plott the Confusion Matrix of this Project ?

  • @rupendrakrishnaraavi4217
    @rupendrakrishnaraavi4217 3 роки тому +1

    Hi is it possible to train the emotion based model with speech by the above procedure?

  • @AyushGupta-je9kn
    @AyushGupta-je9kn 3 роки тому +1

    How to trained machine that if sound is this then do this

  • @CharmVibe24
    @CharmVibe24 5 місяців тому

    What to do when the data is imbalance?

  • @mayurpardeshi395
    @mayurpardeshi395 3 роки тому +1

    This will be end to end project ??

  • @amit_tiger63
    @amit_tiger63 11 місяців тому

    If I want to create real time project like this then how to create its metadata.

  • @rujassohi
    @rujassohi 2 роки тому

    for me, the wav_sample_rate for scipy is exactly the same as librosa why so?

  • @raidahal-smeheen8385
    @raidahal-smeheen8385 Рік тому

    Sorry, I tried to implement the idea on a special project, but so far the highest accuracy I have achieved is 77%
    How can I increase the accuracy

  • @yohannesayana9456
    @yohannesayana9456 2 роки тому

    You're the most selfless guy I have ever seen...Can't wait to see your speech to text tho

  • @paulasam2303
    @paulasam2303 3 роки тому

    I have install librosa successfully but getting an error in "loading audio file with librosa" inspite of correct file address.
    Expecting help from krish.

  • @navneetsinghtaneja5002
    @navneetsinghtaneja5002 2 роки тому

    Sir i have a question, in my mind due to voice deep learning can we interact with animals

  • @suryabolumalla2199
    @suryabolumalla2199 3 роки тому

    Dear sir, can you please help with the vowel sounds and lung disease (based on speech) data bases please 🙏

  • @agammaurya15
    @agammaurya15 5 місяців тому

    is this end to end speech recognition project

  • @syedasma6838
    @syedasma6838 Рік тому

    Sir even after adding the file path and extension . wav I'm getting same error I.e no such file or directory.
    Please tell me what to do??

  • @madhuri_gupta_poetry1076
    @madhuri_gupta_poetry1076 2 роки тому

    Thank u so much sir for such a informative and knowledgeable video. After practicing this code i am getting one error. Kindly help me out. Thanks.
    AttributeError Traceback (most recent call last)
    Input In [38], in ()
    1 plt.figure(figsize=(14,5))
    2 data,sample_rate=librosa.load(filename)
    ----> 3 librosa.display.waveplot(data,sr=sample_rate)
    4 ipd.Audio(filename)
    AttributeError: module 'librosa.display' has no attribute 'waveplot'

    • @bring-it-on
      @bring-it-on 2 роки тому +3

      @Madhuri
      plt.figure(figsize=(14,5))
      data,sample_rate=librosa.load(filename)
      librosa.display.waveshow(data,sr=sample_rate)
      ipd.Audio(filename)
      this will help
      waveshow instead of waveplot

  • @aqdasshayat3158
    @aqdasshayat3158 10 місяців тому

    I have a data set downloaded. but i don,t know how to generate metadata file from it as it is used in the video. where do i convert the data set file into meta data .csv file?

  • @ritanovitasari9653
    @ritanovitasari9653 9 місяців тому

    sir, can you explain whether waveplot and waveshow are the same or different? because I use waveplot and the results are error but if I use waveshow the results are successful but the wavenya is different from sir's. can you please explain. what's wrong why my jupyter doesn't read waveplot.

  • @gayashandulanjana4025
    @gayashandulanjana4025 11 місяців тому

    I have a different voice sound set of human emotions in 6 folders. how can I create the CSV file ?.

  • @SobayoAbiola-ug4tw
    @SobayoAbiola-ug4tw Рік тому

    Krish good day, after downloading this audio file, I was unable to open it

  • @imambilqisthi5928
    @imambilqisthi5928 2 роки тому

    sir , what if sample rate using scipy bigger than using librosa ?

  • @Sidex150-g1p
    @Sidex150-g1p Рік тому

    You're the best.

  • @SA-oj3bo
    @SA-oj3bo 2 роки тому

    Hi, for a long time I am searching for a solution that can recognize dog barking and count how many times /day the dog barks. How to do this please? Can work real time or better to recordh and process it later. ( it does not need to be real time but needs to be accurate) Thanks in advance.

  • @RagaIdentification
    @RagaIdentification Рік тому

    what are the fsID, start, end silence and classID in csv file

  • @humphreyrweikiza6047
    @humphreyrweikiza6047 2 роки тому

    suppose i have a single audio file does the the code file_name= os.path..... still apply
    i am havinng a problem in the file name am constantly retting the error that ther is missing audio file but supprisingly it exist in the folder how can i overcome that

  • @wingsinfotech1530
    @wingsinfotech1530 2 роки тому

    Sir, how to read .raw file using python

  • @alaakamal2588
    @alaakamal2588 2 роки тому

    what is the name of the algorithm that you have used?

  • @Pawan-tc2ih
    @Pawan-tc2ih 2 роки тому

    That was the diagram of how light transverse !

  • @mandaraghava9904
    @mandaraghava9904 Рік тому

    Is ultrasound(8K)-6GB is work in jupyter

  • @aayushronghe8228
    @aayushronghe8228 2 роки тому

    hello sir, i want to run a speaker recognition program using ur code but i have a dataset of my own and i dont know how to generate csv file of this manner from it.Plz help me.

  • @MdKamruzzaman-cz2fq
    @MdKamruzzaman-cz2fq Рік тому

    Thank you brother

  • @durgaganesh423
    @durgaganesh423 2 роки тому

    Hi do we possible to find abnormalities in recored file .wav?

  • @hetvipatel4894
    @hetvipatel4894 2 роки тому

    Hii! Do you have any coding that analysis two voice are different or same?

  • @AnkitKumar-dg4hs
    @AnkitKumar-dg4hs 3 роки тому

    When will the second part come?

  • @ivanarakistain3885
    @ivanarakistain3885 3 роки тому

    Can you help to get TinyML for this? I would like to run classification on a microcontroller.

  • @amalanatu8318
    @amalanatu8318 3 роки тому

    hello sir, not able to download the dataset ...in between download gets interrupted. Is there any alternative? can you please help?

  • @fenixchow1
    @fenixchow1 9 місяців тому

    Thanks!

  • @debatradas9268
    @debatradas9268 2 роки тому

    thank you so much

  • @louerleseigneur4532
    @louerleseigneur4532 3 роки тому

    Thanks Krish

  • @saritasable5274
    @saritasable5274 3 роки тому

    Not able to download the dataset. in between getting failed. is there any alternate way to download

  • @lost_soul8711
    @lost_soul8711 2 роки тому

    how to convert our own sound data set to csv file ??..does anybody knows...???????

  • @youtubetimepasser
    @youtubetimepasser 2 роки тому

    Can I know the realtime application

  • @sayantikachakraborty2055
    @sayantikachakraborty2055 3 роки тому

    Sir the dataset that i am working on doesnt have a csv file and just has the audio..How do i go ahead without having any csv file data?

    • @omingole7304
      @omingole7304 3 роки тому

      If your dataset has only audio files, then download them all and save them in a particular folder. Then follow these steps - 1.Go to this site and follow its instructions to create a column of the audio filenames www.howtoexcel.org/tips-and-tricks/how-to-generate-a-list-of-file-names-from-a-folder-without-vba/ :
      2. Then create a column of the labels of the audio files. 3. You will need some data cleaning in Jupyter notebook to eliminate NaN values and renaming the column names before proceeding further.

  • @navneetsinghtaneja5002
    @navneetsinghtaneja5002 2 роки тому

    Means animals voice dataset communicator

  • @asifnadaf5326
    @asifnadaf5326 2 роки тому

    sir not able to download dataset sir
    pls help!!

  • @noumanijaz5353
    @noumanijaz5353 2 роки тому

    i want to implement this coding on multiple audio file that is the Dcase dataset 2017 challenge can anyone please help me in this regards?

  • @paramamukherjee3436
    @paramamukherjee3436 3 роки тому

    If I haven't any CSV file in my dataset then what to do.?... please reply sir 🙏

  • @my_opiniondemocracy6584
    @my_opiniondemocracy6584 2 роки тому

    how did you get the metadata?

  • @bikashpokharel478
    @bikashpokharel478 2 роки тому

    don't use librosa.waveplot in the newest library insted use librosa.display.waveshow

  • @mohammadmohammadi9268
    @mohammadmohammadi9268 7 місяців тому

    Is it possible you share your code ?

  • @noumanijaz5353
    @noumanijaz5353 2 роки тому

    hello guys anyone please help in implementing DCASE 2017 challenge base line ...

  • @prateek2987singh
    @prateek2987singh 3 роки тому

    facing this issue .... No module named 'librosa'

  • @m.muhtashim1247
    @m.muhtashim1247 3 роки тому +2

    First 😋

  • @pythonhelper9098
    @pythonhelper9098 3 роки тому

    Ipd not defind

  • @beyzaa81
    @beyzaa81 2 роки тому

    is it a CNN?

  • @iftikhar58
    @iftikhar58 2 роки тому

    love from pakistan

  • @mdakramkhan166
    @mdakramkhan166 3 роки тому +3

    Second comment 😅

  • @RagaIdentification
    @RagaIdentification Рік тому

    @krishnaik06 ive created a data set for carnatic music but how do we create a csv file for the dataset

  • @t.bmusic8957
    @t.bmusic8957 Рік тому

    Thank you sir. I learned lot of thing from you.🫀🫀🫀

  • @vt9848
    @vt9848 3 роки тому

    Hai, The urbansound8k dataset has been downloaded as 'Urbansound8k.tar.gz' can anyone tell me how can I do it as a zip file in windows 10? Thanks in advance