Overfitting on Common Voice

Deep Speech
#1

The command I am running:

./DeepSpeech.py --train_files …
–dev_files …/dev.csv --test_files …/test.csv
–checkpoint_dir … --export_dir … --epochs 20 --train_batch_size 64 --dev_batch_size 64 --test_batch_size 64 --early_stop False --summary_dir … --es_steps 50 --learning_rate 0.000075 --dropout_rate 0.20

The model is overfitting, validation error falls max to 70 and the training error continues to decrease. I am thinking of changing the hidden layers. from 2048 to 1800. I have played with the learning rate/dropout and it does not make a difference.

I am training on a p2.xlarge on the Common Voice Dataset.

#2

That does not tell us which language ?

#3

It is the Eng dataset.

#4

That’s not going to be enough data to get you anything really useful. I’d say stop the training at the epoch where you start drifting between training and validation loss ? I don’t really understand what you are asking here.

#5

I am trying to build a Speech to Text model using DeepSpeech. I tried training on the Common Voice dataset from scratch but the accuracy is not good. Are you saying I need more data? Should I train further on the pre-trained model?

#6

Common Voice is the only dataset explored yet.

#7

That’s not really something actionnable.

Yes.

v0.6 model is really getting close now, and it should include currently-released Common Voice English. So maybe you would like to wait … ?

#8

Can you point to existing datasets?
And when is the 0.6 version releasing?

#9

Have you had a look at the Common Voice dataset page ? https://voice.mozilla.org/datasets

I don’t know. Soon, for sure.

1 Like