I am trying to decode a wav file after training a model from scratch, i finished the training and the testing phase without errors and I get the WER & CER and Loss values. NB: I’ve already done the decoding with this command with no errors but it appeared when I tested with a new model.
native_client/deepspeech --model /home/xyz/DeepSpeech/data/exprt_dir/output_graph.pb --scorer /home/xyz/DeepSpeech/data/lm/lm.scorer --audio data/test5.wav --beam_width 9000 > data/decoding.txt
TensorFlow: v2.2.0-15-g518c1d0
DeepSpeech: v0.9.0-alpha.3-0-g78ae08c
Warning: reading entire model file into memory. Transform model file into an mmapped graph to reduce heap usage.
2020-09-22 12:15:00.293543: I tensorflow/core/platform/cpu_feature_guard.cc:143] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
terminate called after throwing an instance of 'std::length_error'
what(): vector::_M_default_append
Abandon (core dumped)`
I tried the command without the scorer, pipe and beam and with a file from the train dataset but the error still persists.
native_client/deepspeech --model /home/xyz/DeepSpeech/data/exprt_dir/output_graph.pb --audio data/decoding_online/test5.wav
TensorFlow: v2.2.0-15-g518c1d0
DeepSpeech: v0.9.0-alpha.3-0-g78ae08c
Warning: reading entire model file into memory. Transform model file into an mmapped graph to reduce heap usage.
2020-09-22 13:56:17.511266: I tensorflow/core/platform/cpu_feature_guard.cc:143] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
terminate called after throwing an instance of 'std::length_error'
what(): vector::_M_default_append
Abandon (core dumped)
you find below the versions of packages i have (running pip list shows more than these packages but i just kept the important ones)
(deepspeech-venv) (base) root@xyz:/home/xyz/DeepSpeech# pip list
Package Version Location
-------------------- ------------ ------------------------------------
deepspeech-training 0.9.0a3 /home/xyz/DeepSpeech/training
ds-ctcdecoder 0.9.0a3
tensorboard 1.15.0
tensorflow-estimator 1.15.1
tensorflow-gpu 1.15.2
Just for future reference, the problem is discussed in length on Discourse.
In general, don't use the master branch if you are not actively developing DeepSpeech as it sometimes breaks.