GitHub Actions for macOS

[lissyx]

No description provided.

[reuben]

Could you push again to make this run?

[reuben]

Some comments but generally looks very good 🚀

[lissyx]

Could you push again to make this run?

You mean to have it running against mozilla/DeepSpeech repo? I think it would run until we merge, but you can check on my fork :)

[reuben]

Could you push again to make this run?

You mean to have it running against mozilla/DeepSpeech repo? I think it would run until we merge, but you can check on my fork :)

Yes. I enabled Actions on this repo. I think instead of working on a giant "GitHub Actions" PR we should rename this to "macOS GitHub Actions" and land ASAP and keep working in separate PRs. And for that we need to get it running here.

[lissyx]

Could you push again to make this run?

You mean to have it running against mozilla/DeepSpeech repo? I think it would run until we merge, but you can check on my fork :)

Yes. I enabled Actions on this repo. I think instead of working on a giant "GitHub Actions" PR we should rename this to "macOS GitHub Actions" and land ASAP and keep working in separate PRs. And for that we need to get it running here.

Sure, that was my goal as well

[lissyx]

Manual testing of the TF / TFLite NodeJS (NodeVS v15):

ds-tc-worker:test-github-actions ds-worker-admin$ PATH=$(pwd)/node-v15.12.0-darwin-x64/bin/:$PATH ./tflite/node_modules/.bin/deepspeech --model deepspeech-0.9.3-models.tflite --scorer deepspeech-0.9.3-models.scorer --audio audio/4507-16021-0012.wav --extended --stream
Loading model from file deepspeech-0.9.3-models.tflite
TensorFlow: v2.3.0-6-g23ad988fcd
DeepSpeech: v0.10.0-alpha.3-57-g0b30a4d8
Loaded model in 0.005774s.
Loading scorer from file deepspeech-0.9.3-models.scorer
Loaded scorer in 0.0004919s.
intermediate:
intermediate:
intermediate:
intermediate:
intermediate: why
intermediate: why should i
intermediate: why should one
intermediate: why should one
intermediate: why should one halt to
intermediate: why should one halt on the
intermediate: why should one halt on the
why should one halt on the way
ds-tc-worker:test-github-actions ds-worker-admin$ PATH=$(pwd)/node-v15.12.0-darwin-x64/bin/:$PATH ./tf/node_modules/.bin/deepspeech --model deepspeech-0.9.3-models.pbmm --scorer deepspeech-0.9.3-models.scorer --audio audio/4507-16021-0012.wav --extended --stream
Loading model from file deepspeech-0.9.3-models.pbmm
TensorFlow: v2.3.0-6-g23ad988fcd
DeepSpeech: v0.10.0-alpha.3-57-g0b30a4d8
2021-03-24 15:00:28.072326: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN)to use the following CPU instructions in performance-critical operations:  AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
Loaded model in 0.01582s.
Loading scorer from file deepspeech-0.9.3-models.scorer
Loaded scorer in 0.0004769s.
intermediate:
intermediate:
intermediate:
intermediate:
intermediate: why s
intermediate: why should i
intermediate: why should one
intermediate: why should one
intermediate: why should one halt
intermediate: why should one halt on the
intermediate: why should one halt on the
why should one halt on the way
ds-tc-worker:test-github-actions ds-worker-admin$

[lissyx]

Manual testing of the TF / TFLite Python (Python 3.9 Homebrew):

ds-tc-worker:test-github-actions ds-worker-admin$ deepspeech --model deepspeech-0.9.3-models.pbmm --scorer deepspeech-0.9.3-models.scorer --audio audio/4507-16021-0012.wav --extended
Loading model from file deepspeech-0.9.3-models.pbmm
TensorFlow: v2.3.0-6-g23ad988fcd
DeepSpeech: v0.10.0-alpha.3-57-g0b30a4d8
2021-03-24 15:03:12.295583: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN)to use the following CPU instructions in performance-critical operations:  AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
Loaded model in 0.014s.
Loading scorer from files deepspeech-0.9.3-models.scorer
Loaded scorer in 0.000314s.
Running inference.
why should one halt on the way
Inference took 1.329s for 2.735s audio file.
ds-tc-worker:test-github-actions ds-worker-admin$ deepspeech --model deepspeech-0.9.3-models.tflite --scorer deepspeech-0.9.3-models.scorer --audio audio/4507-16021-0012.wav --extended
Loading model from file deepspeech-0.9.3-models.tflite
TensorFlow: v2.3.0-6-g23ad988fcd
DeepSpeech: v0.10.0-alpha.3-57-g0b30a4d8
Warning: reading entire model file into memory. Transform model file into an mmapped graph to reduce heap usage.
2021-03-24 15:03:26.890383: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN)to use the following CPU instructions in performance-critical operations:  AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
Data loss: Can't parse deepspeech-0.9.3-models.tflite as binary proto
Traceback (most recent call last):
  File "/usr/local/bin/deepspeech", line 8, in <module>
    sys.exit(main())
  File "/usr/local/lib/python3.9/site-packages/deepspeech/client.py", line 119, in main
    ds = Model(args.model)
  File "/usr/local/lib/python3.9/site-packages/deepspeech/__init__.py", line 38, in __init__
    raise RuntimeError("CreateModel failed with '{}' (0x{:X})".format(deepspeech.impl.ErrorCodeToErrorMessage(status),status))
RuntimeError: CreateModel failed with 'Error reading the proto buffer model file.' (0x3005)
ds-tc-worker:test-github-actions ds-worker-admin$ pip3 install --upgrade deepspeech_tflite-0.10.0a3-cp39-cp39-macosx_10_9_x86_64.whl
Processing ./deepspeech_tflite-0.10.0a3-cp39-cp39-macosx_10_9_x86_64.whl
Requirement already satisfied: numpy in /usr/local/lib/python3.9/site-packages (from deepspeech-tflite==0.10.0a3) (1.20.1)
Installing collected packages: deepspeech-tflite
Successfully installed deepspeech-tflite-0.10.0a3
ds-tc-worker:test-github-actions ds-worker-admin$ deepspeech --model deepspeech-0.9.3-models.tflite --scorer deepspeech-0.9.3-models.scorer --audio audio/4507-16021-0012.wav --extended
Loading model from file deepspeech-0.9.3-models.tflite
TensorFlow: v2.3.0-6-g23ad988fcd
DeepSpeech: v0.10.0-alpha.3-57-g0b30a4d8
Loaded model in 0.0011s.
Loading scorer from files deepspeech-0.9.3-models.scorer
Loaded scorer in 0.0004s.
Running inference.
why should one halt on the way
Inference took 0.585s for 2.735s audio file.
ds-tc-worker:test-github-actions ds-worker-admin$

[lissyx]

@reuben Ready to merge on my side.

[reuben]

Just one comment