Anaconda: packages are not available from current channels

(First try) Trying to install deepspeech with Anaconda on Win10. Package not found.

Command:

C:\Users\jande>conda install -n deepspeechenv deepspeech-gpu

Error:

PackagesNotFoundError: The following packages are not available from current channels:

  - deepspeech-gpu

Current channels:

  - https://repo.anaconda.com/pkgs/main/win-64
  - https://repo.anaconda.com/pkgs/main/noarch
  - https://repo.anaconda.com/pkgs/r/win-64
  - https://repo.anaconda.com/pkgs/r/noarch
  - https://repo.anaconda.com/pkgs/msys2/win-64
  - https://repo.anaconda.com/pkgs/msys2/noarch

How to fix this? Which channel should I add with conda config --append channels ???

Or shoud I first download the zip-file from Github and then what?

No idea, we don’t use / support those. Our packages are available on Pypi.

Thanks. Using pip install deepspeech-gpu from the virtual environment seems to work. One dependency to fix.

 (deepspeechenv) C:\Users\jande>pip install deepspeech-gpu
Collecting deepspeech-gpu
  Downloading https://files.pythonhosted.org/packages/1b/fe/4b57081cd2fc271d9a4bb5784f85646eb4d1c85b3baa3b0c0f583172cb41/deepspeech_gpu-0.6.1-cp37-cp37m-win_amd64.whl (17.3MB)
     |β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 17.3MB 6.4MB/s
Collecting numpy<=1.17.0,>=1.14.5
  Downloading https://files.pythonhosted.org/packages/26/26/73ba03b2206371cdef62afebb877e9ba90a1f0dc3d9de22680a3970f5a50/numpy-1.17.0-cp37-cp37m-win_amd64.whl (12.8MB)
     |β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 12.8MB 6.4MB/s
ERROR: tensorflow 1.14.0 requires google-pasta>=0.1.6, which is not installed.
Installing collected packages: numpy, deepspeech-gpu
  Found existing installation: numpy 1.18.1
    Uninstalling numpy-1.18.1:
      Successfully uninstalled numpy-1.18.1
Successfully installed deepspeech-gpu-0.6.1 numpy-1.17.0

You don’t need tensorflow when you use deepspeech or deepspeech-gpu

That dependency comes from TensorFlow, not deepspeech.

Tensorflow: ah… I thought I did. Ok, no harm done.
Google-Pasta: understood, mo problem. Installed anyway.

Success!

Following the Python example here: Using a pretrained model

(deepspeechenv) F:\Data\Essential\Programming\Python\Deepspeech>deepspeech --model deepspeech-0.6.1-models/output_graph.pbmm --lm deepspeech-0.6.1-models/lm.binary --trie deepspeech-0.6.1-models/trie --audio audio/8455-210777-0068.wav
Loading model from file deepspeech-0.6.1-models/output_graph.pbmm
TensorFlow: v1.14.0-21-ge77504ac6b
DeepSpeech: v0.6.1-0-g3df20fe
2020-01-17 16:50:40.625288: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library nvcuda.dll
2020-01-17 16:50:40.664045: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1640] Found device 0 with properties:
name: GeForce GTX 960 major: 5 minor: 2 memoryClockRate(GHz): 1.304
pciBusID: 0000:01:00.0
2020-01-17 16:50:40.670200: I tensorflow/stream_executor/platform/default/dlopen_checker_stub.cc:25] GPU libraries are statically linked, skip dlopen check.
2020-01-17 16:50:40.676451: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1763] Adding visible gpu devices: 0
2020-01-17 16:52:02.677972: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1181] Device interconnect StreamExecutor with strength 1 edge matrix:
2020-01-17 16:52:02.682473: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1187]      0
2020-01-17 16:52:02.685330: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1200] 0:   N
2020-01-17 16:52:02.689219: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1326] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 3082 MB memory) -> physical GPU (device: 0, name: GeForce GTX 960, pci bus id: 0000:01:00.0, compute capability: 5.2)
Loaded model in 82.4s.
Loading language model from files deepspeech-0.6.1-models/lm.binary deepspeech-0.6.1-models/trie
Loaded language model in 0.0218s.
Running inference.
your power is sufficient i said
Inference took 1.423s for 2.590s audio file.

(deepspeechenv) F:\Data\Essential\Programming\Python\Deepspeech>

1 Like

Thanks, so it means it’s working properly under conda on Windows.

Yes it works under Conda on WIndows 10 using a low entry GPU.

1 Like

I’m mentionning this as a good news, because in the past we have got reports of unexpected behavior on that kind of setup.

Ok. Mind you, I ran one sample and that works. Hope this continues to be the case. So far, so good.

Just got the mic_vad_streaming working on Win 10, Anaconda, Visual Studio 2019, Python 3.7.6.

Portaudio install using: conda install -c anaconda portaudio
followed by conda install pyaudio