mirror of
https://github.com/TaterTotterson/microWakeWord-Trainer-Nvidia-Docker.git
synced 2026-06-12 20:10:19 -06:00
cb81f7f02d
The files in the `cli` directory allow you to train wake words from the command line without needing to use the Jupyter notebook or a web browser. Basically, the logic from the notebook has been placed in separate shell scripts and python files wrapped by 3 high-level scripts that do the following: * setup_python_venv: Creates a Python virtual environment with all the packages needed to train. The venv is created in the container's /data directory and is therefore stored on the host, not in the container's root docker volume. * setup_training_datasets: Downloads, extracts and converts the MIT RIR, FMA, Audioset and Negative training reference datasets. Also stored in /data. * train_wake_word: Generates the wake word samples, augments them with the audio from the training datasets, and finally runs the microwakeword training. The resulting model tflite and json files are placed in the /data/output directory. See the README.md file for much more information.
130 lines
3.0 KiB
Bash
Executable File
130 lines
3.0 KiB
Bash
Executable File
#!/bin/bash
|
|
|
|
PROGPATH=$(realpath "$0")
|
|
PROGDIR=$(dirname "${PROGPATH}")
|
|
TRAINING_STEPS=40000
|
|
DATA_DIR=/data
|
|
source "${PROGDIR}/shell.functions"
|
|
|
|
source "${DATA_DIR}/.venv/bin/activate"
|
|
|
|
export TF_CPP_MIN_LOG_LEVEL=9
|
|
export GLOG_minloglevel=2
|
|
export GRPC_VERBOSITY="ERROR"
|
|
|
|
echo -e "\n===== Testing Python Environment =====\n"
|
|
|
|
echo -e "\n===== Testing Cuda =====\n"
|
|
"${PROGDIR}/cudainfo"
|
|
|
|
python - 2>/dev/null <<EOF
|
|
import os, sys
|
|
|
|
print("\n===== Testing Tensorflow =====\n")
|
|
try:
|
|
from ai_edge_litert.interpreter import Interpreter
|
|
import tensorflow as tf
|
|
|
|
try:
|
|
with tf.device("/GPU:0"):
|
|
a = tf.random.normal([10000, 10000])
|
|
b = tf.random.normal([10000, 10000])
|
|
c = tf.matmul(a, b)
|
|
if c.device.find("GPU") >= 0:
|
|
result = "Available - " + c.device
|
|
else:
|
|
result = "Not available"
|
|
except:
|
|
result = "Not available"
|
|
|
|
print("GPU:", result)
|
|
|
|
try:
|
|
with tf.device("/CPU:0"):
|
|
a = tf.random.normal([10000, 10000])
|
|
b = tf.random.normal([10000, 10000])
|
|
c = tf.matmul(a, b)
|
|
result = "Available - " + c.device
|
|
except:
|
|
result = "Not available"
|
|
|
|
print("CPU:", result)
|
|
except:
|
|
print("Tensorflow not available")
|
|
EOF
|
|
|
|
|
|
python - 2>/dev/null <<EOF
|
|
import os, sys
|
|
print("\n===== Testing Torch =====\n")
|
|
|
|
try:
|
|
import torch
|
|
|
|
if torch.cuda.is_available():
|
|
print(f"GPU: Available - {torch.cuda.get_device_name(0)}")
|
|
else:
|
|
print("GPU:", "Not available")
|
|
print("CPU:", "Available")
|
|
except:
|
|
print("Torch not available")
|
|
EOF
|
|
|
|
python - 2>/dev/null <<EOF
|
|
import os, sys
|
|
print("\n===== Testing onnxruntime =====\n")
|
|
|
|
try:
|
|
import onnxruntime as ort
|
|
|
|
providers = ort.get_available_providers()
|
|
if 'CUDAExecutionProvider' in providers:
|
|
print("GPU:", "Available")
|
|
else:
|
|
print("GPU:", "Not available")
|
|
|
|
if 'CPUExecutionProvider' in providers:
|
|
print("CPU:", "Available")
|
|
else:
|
|
print("CPU:", "Not available")
|
|
|
|
if 'TensorrtExecutionProvider' in providers:
|
|
print("TensorRT:", "Available")
|
|
else:
|
|
print("TensorRT:", "Not available")
|
|
except:
|
|
print("onnxruntime not available")
|
|
EOF
|
|
|
|
python - 2>/dev/null <<EOF
|
|
import os, sys
|
|
|
|
print("\n===== Testing micro-wake-word =====\n")
|
|
|
|
try:
|
|
import numpy as np
|
|
import librosa
|
|
from mmap_ninja.ragged import RaggedMmap
|
|
from microwakeword.audio.augmentation import Augmentation
|
|
from microwakeword.audio.clips import Clips
|
|
from microwakeword.audio.spectrograms import SpectrogramGeneration
|
|
from microwakeword.audio.audio_utils import save_clip
|
|
|
|
print("micro-wake-word available")
|
|
except:
|
|
print("micro-wake-word not available")
|
|
|
|
print("")
|
|
EOF
|
|
|
|
echo -e "===== Testing piper-sample-generator =====\n"
|
|
|
|
./tools/piper-sample-generator/generate_samples.py --help &>/dev/null && {
|
|
echo "piper-sample-generator available"
|
|
} || {
|
|
echo "piper-sample-generator not available"
|
|
}
|
|
|
|
echo
|
|
echo -e "\n===== Python Environment Testing Complete =====\n"
|