search

Prachi98-palvi / Spoken_Indian_language_Identification_using_CNN

Audio identification, by converting audio files into spectrograms and then building an image identification CNN model is the base of this project.
2 stars 0 forks source link

how to load dataset #1

Open IRSHAD500THUKROO opened 3 years ago

IRSHAD500THUKROO commented 3 years ago

mam could u please tell me the what the /content/drive/My Drive/audio_files contains , whether it contains all the .wav files of 7 languages or the all unzip files .Please tell me how to load data exactly . Thanks for anticipatio

Prachi98-palvi commented 3 years ago

Hello, The drive contains all .wav files. The audio files were converted into image format. Specifically spectrograms. All those spectrogram images were used as an input data to train the CNN model (60-20-20 - train-test-validate)

Regards, Prachi

On Wed, Apr 7, 2021, 4:11 PM IRSHAD500THUKROO @.***> wrote:

mam could u please tell me the what the /content/drive/My Drive/audio_files contains , whether it contains all the .wav files of 7 languages or the all unzip files .Please tell me how to load data exactly . Thanks for anticipatio

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/Prachi98-palvi/Spoken_Indian_language_Identification_using_CNN/issues/1, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOSBI3PDELAIVP7EHRFDSJDTHQZF7ANCNFSM42QQ55TA .

IRSHAD500THUKROO commented 3 years ago

thanks a lot mam, it means a lot to me. mam if i need further help can i refer to u. Will this model work on kashmiri and english language?

On Wed, Apr 7, 2021 at 4:17 PM PrachiPalvi @.***> wrote:

Hello, The drive contains all .wav files. The audio files were converted into image format. Specifically spectrograms. All those spectrogram images were used as an input data to train the CNN model (60-20-20 - train-test-validate)

Regards, Prachi

On Wed, Apr 7, 2021, 4:11 PM IRSHAD500THUKROO @.***> wrote:

mam could u please tell me the what the /content/drive/My Drive/audio_files contains , whether it contains all the .wav files of 7 languages or the all unzip files .Please tell me how to load data exactly . Thanks for anticipatio

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub < https://github.com/Prachi98-palvi/Spoken_Indian_language_Identification_using_CNN/issues/1 , or unsubscribe < https://github.com/notifications/unsubscribe-auth/AOSBI3PDELAIVP7EHRFDSJDTHQZF7ANCNFSM42QQ55TA

.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/Prachi98-palvi/Spoken_Indian_language_Identification_using_CNN/issues/1#issuecomment-814810841, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMEKQRLQSUGPY4EP6XZ7VTLTHQZ3FANCNFSM42QQ55TA .

Prachi98-palvi commented 3 years ago

Sure. It will work for all languages. The limitation to this model is the files user as input data are noise free, hence making it's real life application equal to negligible.

I would recommend to explore that part, if you have noise in your data.

On Wed, Apr 7, 2021, 8:21 PM IRSHAD500THUKROO @.***> wrote:

thanks a lot mam, it means a lot to me. mam if i need further help can i refer to u. Will this model work on kashmiri and english language?

On Wed, Apr 7, 2021 at 4:17 PM PrachiPalvi @.***> wrote:

Hello, The drive contains all .wav files. The audio files were converted into image format. Specifically spectrograms. All those spectrogram images were used as an input data to train the CNN model (60-20-20 - train-test-validate)

Regards, Prachi

On Wed, Apr 7, 2021, 4:11 PM IRSHAD500THUKROO @.***> wrote:

mam could u please tell me the what the /content/drive/My Drive/audio_files contains , whether it contains all the .wav files of 7 languages or the all unzip files .Please tell me how to load data exactly . Thanks for anticipatio

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <

https://github.com/Prachi98-palvi/Spoken_Indian_language_Identification_using_CNN/issues/1

, or unsubscribe <

https://github.com/notifications/unsubscribe-auth/AOSBI3PDELAIVP7EHRFDSJDTHQZF7ANCNFSM42QQ55TA

.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub < https://github.com/Prachi98-palvi/Spoken_Indian_language_Identification_using_CNN/issues/1#issuecomment-814810841 , or unsubscribe < https://github.com/notifications/unsubscribe-auth/AMEKQRLQSUGPY4EP6XZ7VTLTHQZ3FANCNFSM42QQ55TA

.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Prachi98-palvi/Spoken_Indian_language_Identification_using_CNN/issues/1#issuecomment-814978980, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOSBI3M6FQ4JDJ25MGECM2DTHRWPNANCNFSM42QQ55TA .

IRSHAD500THUKROO commented 3 years ago

Respected mam

I tried to test the model by using three languages (bengali , marathi and tamil) each of 100 files . When i execute the model it shows me a warning and the results were like this (image shows the result). warning (/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/sequential.py:450: UserWarning: model.predict_classes() is deprecated and will be removed after 2021-01-01. Please use instead: np.argmax(model.predict(x), axis=-1), if your model does multi-class classification (e.g. if it uses a softmax last-layer activation). (model.predict(x) > 0.5).astype("int32"), if your model does binary classification (e.g. if it uses a sigmoid last-layer activation). warnings.warn('model.predict_classes() is deprecated and ')

mam could you explain the working , Please Is it possible i can contact u mam.

Regards Irshad ahmad

Prachi98-palvi commented 3 years ago

Hi Irshad, I think you can reach out to Sidhant Mankar. You may reach out to him on linkedin and see if he is willing to help you on this.

Thanks and Regards Prachi Palvi

On Thu, 22 Apr 2021 at 12:10, IRSHAD500THUKROO @.***> wrote:

Respected mam

I tried to test the model by using three languages (bengali , marathi and tamil) each of 100 files . When i execute the model it shows me a warning and the results were like this (image shows the result). warning (/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/engine/sequential.py:450: UserWarning: model.predict_classes() is deprecated and will be removed after 2021-01-01. Please use instead: np.argmax(model.predict(x), axis=-1), if your model does multi-class classification (e.g. if it uses a softmax last-layer activation). (model.predict(x) > 0.5).astype("int32"), if your model does binary classification (e.g. if it uses a sigmoid last-layer activation). warnings.warn('model.predict_classes() is deprecated and ')

mam could you explain the working , Please Is it possible i can contact u mam.

Regards Irshad ahmad

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Prachi98-palvi/Spoken_Indian_language_Identification_using_CNN/issues/1#issuecomment-824581883, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOSBI3LMYM3B6Z66NNTN4BDTJ7AFBANCNFSM42QQ55TA .