wav2letter

wav2letter copied to clipboard

Libri-1K training

23

Hello, I am trying to run Libri-1K transformer training. After 20 epoch train-WER=327 (please see log below). Is something wrong with my running? I0530 04:41:50.446601 981 Train.cpp:573] Epoch 20 started!...

AlexandderGorodetski

which wer cat you get when train streaming_convnets using librispeech only 1k hours data?

4

I try to reproduce the streaming_convnets in librispeech data on a 4-gpu machine, I find it hard to train all the data with libri-light, so I just use 1k hours...

qzfnihao

Past hypothesis rescoring based on future context with beam search

4

When I run online inference, it seems like once a part of a sentence is printed to the stdout, the part of the sentence never changes regardless of the following...

janvainer

Stuck when running on multi-GPU

19

### Question Hi, could someone help me why my train is stucked when using multi-gpu? #### Additional Context   Stuck, no training iteration done (I am using 1 iteration...

tryanbot

Issue on model fine tuning:sota/2019/librispeech

4

Hi, I'm using the fork command on am_resnet_ctc_librispeech_dev_other.bin to adapt the model to my own dataset, and i got the following errors which says `Loss has NaN values.` ``` I0723...

Rootian

Help to solve training error

9

Hello, I was training a new model. But after around 100000 iterations (maybe still in the first epoch), I got the fallowing error message. *** Aborted at 1590963336 (unix time)...

Hui-Ling

Data preparation for Language Modeling

20

I have some questions regarding language model. 1. I want to build a character level LM. Previously I built a word level LM using KENLM and for that I created...

samin9796

Pre-trained models

1

### Question Are there any other wav2letter models published besides the one in AWS S3? http://dl.fbaipublicfiles.com/wav2letter/inference/examples/model Thanks

latha-ai

Do you have any plan to realize mwer criterion?

4

HI, as we know , mwer can reduce wer by 8% more on ctc and s2s. Does it have any play to realize it?

qzfnihao

Help: OOM error (ArrayFire exception) during training a tds-s2s model

6

Hello, While training a TDS seq-to-seq model, (CUDA) OOM errors (ArrayFire exceptions) are raised constantly. Training configuration is: - 5k hours of training data - DGX-1v (8 x V100, each...

dh7ahn