Assert channels == 1 # only mono supported for now AssertionError

DeepSpeech
#1

i am trying out fine tuning and prepared own data of just 2-3 audios to check.
i am getting this error

File “/content/DeepSpeech/training/deepspeech_training/util/audio.py”, line 367, in pcm_to_np
** assert channels == 1 # only mono supported for now**
AssertionError
"""

#2

You have to convert your audio-samples to mono first. You could use ffmpeg or sox for this.

#3

Thank you. I used sox to convert it to mono. :slight_smile: It worked now