Google Colab notebook needs update

[BillBBBB]

The colab notebook seems to need some updates to its dependencies.

---

ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
openwakeword 0.6.0 requires speexdsp-ns<1,>=0.1.2; platform_system == "Linux", which is not installed.
openwakeword 0.6.0 requires ai-edge-litert<3,>=2.0.2; platform_system == "Linux" or platform_system == "Darwin", but you have ai-edge-litert 1.4.0 which is incompatible.
onnx2tf 2.4.0 requires ai-edge-litert==2.1.2, but you have ai-edge-litert 1.4.0 which is incompatible.
onnx2tf 2.4.0 requires onnx==1.20.1, but you have onnx 1.19.1 which is incompatible.
onnx2tf 2.4.0 requires onnxruntime==1.24.3, but you have onnxruntime 1.22.1 which is incompatible.

ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
openwakeword 0.6.0 requires speexdsp-ns<1,>=0.1.2; platform_system == "Linux", which is not installed.
Successfully installed ai-edge-litert-2.1.2 onnx-1.20.1 onnxruntime-1.24.3
Collecting onnx==1.19.1
Using cached onnx-1.19.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (7.0 kB)
Requirement already satisfied: numpy>=1.22 in /usr/local/lib/python3.12/dist-packages (from onnx==1.19.1) (1.26.4)
Requirement already satisfied: protobuf>=4.25.1 in /usr/local/lib/python3.12/dist-packages (from onnx==1.19.1) (4.25.5)
Requirement already satisfied: typing_extensions>=4.7.1 in /usr/local/lib/python3.12/dist-packages (from onnx==1.19.1) (4.15.0)
Requirement already satisfied: ml_dtypes>=0.5.0 in /usr/local/lib/python3.12/dist-packages (from onnx==1.19.1) (0.5.1)
Using cached onnx-1.19.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (18.2 MB)
Installing collected packages: onnx
Attempting uninstall: onnx
Found existing installation: onnx 1.20.1
Uninstalling onnx-1.20.1:
Successfully uninstalled onnx-1.20.1
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
onnx2tf 2.4.0 requires onnx==1.20.1, but you have onnx 1.19.1 which is incompatible.

---

---

ValueError Traceback (most recent call last)

/tmp/ipykernel_23700/3645593403.py in <cell line: 0>()
74 import uuid
75 import yaml
---> 76 import datasets
77 import scipy
78 from tqdm import tqdm

8 frames

/usr/local/lib/python3.12/dist-packages/numpy/random/_pickle.py in
----> 1 from .mtrand import RandomState
2 from ._philox import Philox
3 from ._pcg64 import PCG64, PCG64DXSM
4 from ._sfc64 import SFC64
5

numpy/random/mtrand.pyx in init numpy.random.mtrand()

ValueError: numpy.dtype size changed, may indicate binary incompatibility. Expected 96 from C header, got 88 from PyObject

[BillBBBB]

Also a huge thank you for maintaining this I am really looking forward to getting this up and working!

[muhammadabidPro]

🤫:https://colab.research.google.com/github/arcosoph/nanowakeword/blob/main/notebooks/Train_Your_First_Wake_Word_Model.ipynb

[hoho2312]

I just modified the dependencies of the official colab file. Feel free to try.

automatic_model_training_simple_modified.ipynb

[muhammadabidPro]

😑 Train_Your_First_Wake_Word_Model.ipynb

Top 1 😑

[BillBBBB]

I just modified the dependencies of the official colab file. Feel free to try.

automatic_model_training_simple_modified.ipynb
Seems to be working thank you!

[muhammadabidPro]

But what's your problem with trying it out?

😑 Train_Your_First_Wake_Word_Model.ipynb

You will find everything here. ༼ つ ◕_◕ ༽つ Try Now!

[POBOROJO]

I just modified the dependencies of the official colab file. Feel free to try.

automatic_model_training_simple_modified.ipynb

i was training a wake model but the accuracy was 70 to 75 around how are you getting 88 ? is it because of the updated dependencies used ?

[muhammadabidPro]

All problems one solution

• Train_Your_First_Wake_Word_Model.ipynb

[jclsn]

I just modified the dependencies of the official colab file. Feel free to try.

automatic_model_training_simple_modified.ipynb

Getting errors as well

--------------------------------------------------------------------------

UnpicklingError                           Traceback (most recent call last)

[/tmp/ipykernel_609/2056216887.py](https://localhost:8080/#) in <cell line: 0>()
     48                 )
     49 
---> 50 text_to_speech(target_word)
     51 
     52 from IPython.display import Audio

2 frames

[/usr/local/lib/python3.12/dist-packages/torch/serialization.py](https://localhost:8080/#) in load(f, map_location, pickle_module, weights_only, mmap, **pickle_load_args)
   1576                         )
   1577                     except pickle.UnpicklingError as e:
-> 1578                         raise pickle.UnpicklingError(_get_wo_message(str(e))) from None
   1579                 return _load(
   1580                     opened_zipfile,

UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. 
	(1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source.
	(2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message.
	WeightsUnpickler error: Unsupported global: GLOBAL piper_train.vits.models.SynthesizerTrn was not an allowed global by default. Please use `torch.serialization.add_safe_globals([piper_train.vits.models.SynthesizerTrn])` or the `torch.serialization.safe_globals([piper_train.vits.models.SynthesizerTrn])` context manager to allowlist this global if you trust this class/function.

Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html.

[jclsn]

All problems one solution

* [Train_Your_First_Wake_Word_Model.ipynb](https://colab.research.google.com/github/arcosoph/nanowakeword/blob/main/notebooks/Train_Your_First_Wake_Word_Model.ipynb)

Would be nice if there was an easier way to enter the data and a way to prelisten the wakeword as in the original notebook!