deepspeech-german: loading output_graph.tflite deepspeech-tflite result in segmentation fault

When loading the tflite version in deepspeech-tflite version 0.9.2 and also 0.9.0 I get an segmentation fault SIGSEGV. Can this be replicated?

from deepspeech import Model

model = 'output_graph.tflite'
ds = Model(model)

What could be the issue? I do not see any stacktrace or useful information how to find the source of the problem.

About this issue

Original URL
State: closed
Created 4 years ago
Comments: 17 (14 by maintainers)

Most upvoted comments

I will close this one now. I will create a pull request soon for some updates for some scripts and the README. The steps from the README are not working anymore with DeepSpeech 0.9.3. I had also added DeepSpeech and KenLM as submodules, which make the checkout more straight forward and is always using the version which was tested against this project.

kaoh on Jan 10, 2021

@terler188 : Thank you for reminding. I have now uploaded the tflite model to the drive. Thanks to @koh-osug

AASHISHAG on Jan 6, 2021

When re-exporting the tflite model with:

../venv_tools/bin/python DeepSpeech.py --export_tflite --export_dir . --alphabet_config_path=deepspeech-german/alphabet.txt --checkpoint_dir=deepspeech-german

The segmentation fault is gone. Maybe you can contribute to the opened deepspeech issue above and look again into your build chain. Something seems to be not functional or simply the uploaded tflite file is corrupt?

koh-osug on Dec 14, 2020