mirror of
https://github.com/snakers4/silero-vad.git
synced 2026-02-04 17:39:22 +08:00
410 lines
9.3 KiB
Plaintext
410 lines
9.3 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "sVNOuHQQjsrp"
|
|
},
|
|
"source": [
|
|
"# PyTorch Example"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "9ZTzCtc5kYVg"
|
|
},
|
|
"source": [
|
|
"## Install Dependencies"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2020-12-15T14:00:15.701867Z",
|
|
"start_time": "2020-12-15T14:00:09.512876Z"
|
|
},
|
|
"cellView": "form",
|
|
"collapsed": true,
|
|
"id": "rllMjjsekbjt"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"#@title Install and Import Dependencies\n",
|
|
"\n",
|
|
"# this assumes that you have a relevant version of PyTorch installed\n",
|
|
"!pip install -q torchaudio soundfile\n",
|
|
"\n",
|
|
"import glob\n",
|
|
"import torch\n",
|
|
"torch.set_num_threads(1)\n",
|
|
"\n",
|
|
"from IPython.display import Audio\n",
|
|
"from pprint import pprint\n",
|
|
"\n",
|
|
"model, utils = torch.hub.load(repo_or_dir='snakers4/silero-vad',\n",
|
|
" model='silero_vad',\n",
|
|
" force_reload=True)\n",
|
|
"\n",
|
|
"(get_speech_ts,\n",
|
|
" save_audio,\n",
|
|
" read_audio,\n",
|
|
" state_generator,\n",
|
|
" single_audio_stream,\n",
|
|
" collect_speeches) = utils\n",
|
|
"\n",
|
|
"files_dir = torch.hub.get_dir() + '/snakers4_silero-vad_master/files'"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "fXbbaUO3jsrw"
|
|
},
|
|
"source": [
|
|
"## Full Audio"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2020-12-15T13:09:56.879818Z",
|
|
"start_time": "2020-12-15T13:09:56.864765Z"
|
|
},
|
|
"id": "aI_eydBPjsrx"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"wav = read_audio(f'{files_dir}/en.wav')\n",
|
|
"# get speech timestamps from full audio file\n",
|
|
"speech_timestamps = get_speech_ts(wav, model,\n",
|
|
" num_steps=4)\n",
|
|
"pprint(speech_timestamps)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2020-12-15T13:09:58.941063Z",
|
|
"start_time": "2020-12-15T13:09:58.887006Z"
|
|
},
|
|
"id": "OuEobLchjsry"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"# merge all speech chunks to one audio\n",
|
|
"save_audio('only_speech.wav',\n",
|
|
" collect_speeches(speech_timestamps, wav), 16000) \n",
|
|
"Audio('only_speech.wav')"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "iDKQbVr8jsry"
|
|
},
|
|
"source": [
|
|
"## Single Audio Stream"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2020-12-15T13:09:59.199321Z",
|
|
"start_time": "2020-12-15T13:09:59.196823Z"
|
|
},
|
|
"id": "q-lql_2Wjsry"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"wav = f'{files_dir}/en.wav'\n",
|
|
"\n",
|
|
"for batch in single_audio_stream(model, wav):\n",
|
|
" if batch:\n",
|
|
" print(batch)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "KBDVybJCjsrz"
|
|
},
|
|
"source": [
|
|
"## Multiple Audio Streams"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2020-12-15T13:10:03.590358Z",
|
|
"start_time": "2020-12-15T13:10:03.587071Z"
|
|
},
|
|
"id": "BK4tGfWgjsrz"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"audios_for_stream = glob.glob(f'{files_dir}/*.wav')\n",
|
|
"len(audios_for_stream) # total 4 audios"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2020-12-15T13:10:15.762491Z",
|
|
"start_time": "2020-12-15T13:10:03.591388Z"
|
|
},
|
|
"id": "v1l8sam1jsrz"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"for batch in state_generator(model, audios_for_stream, audios_in_stream=2): # 2 audio stream\n",
|
|
" if batch:\n",
|
|
" pprint(batch)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "57avIBd6jsrz"
|
|
},
|
|
"source": [
|
|
"# ONNX Example"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "bL4kn4KJrlyL"
|
|
},
|
|
"source": [
|
|
"## Install Dependencies"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"cellView": "form",
|
|
"id": "Q4QIfSpprnkI"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"#@title Install and Import Dependencies\n",
|
|
"\n",
|
|
"# this assumes that you have a relevant version of PyTorch installed\n",
|
|
"!pip install -q torchaudio soundfile onnxruntime\n",
|
|
"\n",
|
|
"import glob\n",
|
|
"import onnxruntime\n",
|
|
"from pprint import pprint\n",
|
|
"\n",
|
|
"from IPython.display import Audio\n",
|
|
"\n",
|
|
"_, utils = torch.hub.load(repo_or_dir='snakers4/silero-vad',\n",
|
|
" model='silero_vad',\n",
|
|
" force_reload=True)\n",
|
|
"\n",
|
|
"(get_speech_ts,\n",
|
|
" save_audio,\n",
|
|
" read_audio,\n",
|
|
" state_generator,\n",
|
|
" single_audio_stream,\n",
|
|
" collect_speeches) = utils\n",
|
|
"\n",
|
|
"files_dir = torch.hub.get_dir() + '/snakers4_silero-vad_master/files'\n",
|
|
"\n",
|
|
"def init_onnx_model(model_path: str):\n",
|
|
" return onnxruntime.InferenceSession(model_path)\n",
|
|
"\n",
|
|
"def validate_onnx(model, inputs):\n",
|
|
" with torch.no_grad():\n",
|
|
" ort_inputs = {'input': inputs.cpu().numpy()}\n",
|
|
" outs = model.run(None, ort_inputs)\n",
|
|
" outs = [torch.Tensor(x) for x in outs]\n",
|
|
" return outs"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "5JHErdB7jsr0"
|
|
},
|
|
"source": [
|
|
"## Full Audio"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2020-12-15T13:09:06.643812Z",
|
|
"start_time": "2020-12-15T13:09:06.473386Z"
|
|
},
|
|
"id": "krnGoA6Kjsr0"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"model = init_onnx_model(f'{files_dir}/model.onnx')\n",
|
|
"wav = read_audio(f'{files_dir}/en.wav')\n",
|
|
"\n",
|
|
"# get speech timestamps from full audio file\n",
|
|
"speech_timestamps = get_speech_ts(wav, model, num_steps=4, run_function=validate_onnx) \n",
|
|
"pprint(speech_timestamps)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2020-12-15T13:09:08.862421Z",
|
|
"start_time": "2020-12-15T13:09:08.820014Z"
|
|
},
|
|
"id": "B176Lzfnjsr1"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"# merge all speech chunks to one audio\n",
|
|
"save_audio('only_speech.wav', collect_speeches(speech_timestamps, wav), 16000)\n",
|
|
"Audio('only_speech.wav')"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "Rio9W50gjsr1"
|
|
},
|
|
"source": [
|
|
"## Single Audio Stream"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2020-12-15T13:09:09.606031Z",
|
|
"start_time": "2020-12-15T13:09:09.504239Z"
|
|
},
|
|
"id": "IPkl8Yy1jsr1"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"model = init_onnx_model(f'{files_dir}/model.onnx')\n",
|
|
"wav = f'{files_dir}/en.wav'"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2020-12-15T13:09:11.453171Z",
|
|
"start_time": "2020-12-15T13:09:09.633435Z"
|
|
},
|
|
"id": "NC6Jim0hjsr1"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"for batch in single_audio_stream(model, wav, run_function=validate_onnx):\n",
|
|
" if batch:\n",
|
|
" pprint(batch)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "WNZ42u0ajsr1"
|
|
},
|
|
"source": [
|
|
"## Multiple Audio Streams"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2020-12-15T13:09:11.540423Z",
|
|
"start_time": "2020-12-15T13:09:11.455706Z"
|
|
},
|
|
"id": "XjhGQGppjsr1"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"model = init_onnx_model(f'{files_dir}/model.onnx')\n",
|
|
"audios_for_stream = glob.glob(f'{files_dir}/*.wav')\n",
|
|
"pprint(len(audios_for_stream)) # total 4 audios"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"ExecuteTime": {
|
|
"end_time": "2020-12-15T13:09:19.565434Z",
|
|
"start_time": "2020-12-15T13:09:11.552097Z"
|
|
},
|
|
"id": "QI7-arlqjsr2"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"for batch in state_generator(model, audios_for_stream, audios_in_stream=2, run_function=validate_onnx): # 2 audio stream\n",
|
|
" if batch:\n",
|
|
" pprint(batch)"
|
|
]
|
|
}
|
|
],
|
|
"metadata": {
|
|
"colab": {
|
|
"name": "silero-vad.ipynb",
|
|
"provenance": []
|
|
},
|
|
"kernelspec": {
|
|
"display_name": "Python 3",
|
|
"language": "python",
|
|
"name": "python3"
|
|
},
|
|
"language_info": {
|
|
"codemirror_mode": {
|
|
"name": "ipython",
|
|
"version": 3
|
|
},
|
|
"file_extension": ".py",
|
|
"mimetype": "text/x-python",
|
|
"name": "python",
|
|
"nbconvert_exporter": "python",
|
|
"pygments_lexer": "ipython3",
|
|
"version": "3.8.3"
|
|
},
|
|
"toc": {
|
|
"base_numbering": 1,
|
|
"nav_menu": {},
|
|
"number_sections": true,
|
|
"sideBar": true,
|
|
"skip_h1_title": false,
|
|
"title_cell": "Table of Contents",
|
|
"title_sidebar": "Contents",
|
|
"toc_cell": false,
|
|
"toc_position": {},
|
|
"toc_section_display": true,
|
|
"toc_window_display": false
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 0
|
|
}
|