Audio_dataset_from_directory: How to fix it?

Keras
,
#1

hi everyone, recently i started using tf.data.Dataset, when i create Dataset from audio_dataset_from_directory method of keras.utils there is no error. The image below shows the success message.
image

Then I read the shape of the feature and label and got the error message:

InvalidArgumentError: {{function_node _wrapped__IteratorGetNext_output_types_2_device/job:localhost/replica:0/task:0/device:CPU:0}} No data chunk found in WAV
[[{{node DecodeWav}}]] [Op:IteratorGetNext]

I also tested with tf’s mini_speech_commands dataset and it’s fine but when I put the TSP TSP Lab - Data, I got the above message, how to fix it?
Thanks for taking the time to read, :))
Full code below.

image
image936×496 19 KB

image
image1022×337 11.4 KB

image
image1240×390 19.5 KB

#3

Hi @ToHoangMinhTien, Welcome to TensorFlow Forum!

while reproducing the issue in the colab we observed the same. we are working on it. Thank You.