DeepSpeech2 thown error "File contains data in an unimplemented format"

[jacquesqiao]

We downloaded the librespeech model for speech recognition, but soundfile does not seem to be working with the code provided. We have pip installed soundfile and it seems to import just fine, but the methods do not seem to be working. Is there any light that could be shed on our problem?Specifically, https://github.com/PaddlePaddle/models/blob/develop/deep_speech_2/data/librispeech/librispeech.py gives error "File contains data in an unimplemented format." in line 103

[jacquesqiao]

what is the format of your sound file? is it can be convert to a .flac format? use the tool like https://audio.online-convert.com/convert-to-flac

[ganzhong007]

the same problem I meet. anyone can tell me how to resolve it ?

[zhxfl]

the same problem I meet. anyone can tell me how to resolve it ?

[muntasir2000]

I had the same issue. I solved it by installing libsndfile. http://www.mega-nerd.com/libsndfile/files/libsndfile-1.0.28.tar.gz

To install -

tar -zxvf libsndfile-1.0.28.tar.gz
cd libsndfile-1.0.28
./configure > /dev/null && make > /dev/null && make install > /dev/null

[PasaOpasen]

@ganzhong007

the same problem I meet. anyone can tell me how to resolve it ?

have u solved this problem?

[Always-prog]

Hmm. I have a similar problem

[llllaaapp]

have u solved this problem?

[Always-prog]

@llllaaapp I've already forgotten 🙂

[Always-prog]

But as I see @muntasir2000 solved it by installing http://www.mega-nerd.com/libsndfile/files/libsndfile-1.0.28.tar.gz