deepgeek/silero-vad
2
0
Fork 0
mirror of https://github.com/snakers4/silero-vad.git synced 2026-02-05 18:09:22 +08:00
Code Issues Packages Projects Releases Wiki Activity

v5 initial push

Browse Source
This commit is contained in:
adamnsandle
2024-06-27 15:41:20 +00:00
parent 8145ed9a91
commit fd1f1a62b7
4 changed files with 61 additions and 38 deletions
Download Patch File Download Diff File Expand all files Collapse all files

44
silero-vad.ipynb
View File

@@ -46,7 +46,7 @@
"USE_ONNX = False # change this to True if you want to test onnx model\n",
"if USE_ONNX:\n",
" !pip install -q onnxruntime\n",
" \n",
"\n",
"model, utils = torch.hub.load(repo_or_dir='snakers4/silero-vad',\n",
" model='silero_vad',\n",
" force_reload=True,\n",
@@ -65,16 +65,7 @@
"id": "fXbbaUO3jsrw"
},
"source": [
"## Full Audio"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "RAfJPb_a-Auj"
},
"source": [
"**Speech timestapms from full audio**"
"## Speech timestapms from full audio"
]
},
{
@@ -101,10 +92,33 @@
"source": [
"# merge all speech chunks to one audio\n",
"save_audio('only_speech.wav',\n",
" collect_chunks(speech_timestamps, wav), sampling_rate=SAMPLING_RATE) \n",
" collect_chunks(speech_timestamps, wav), sampling_rate=SAMPLING_RATE)\n",
"Audio('only_speech.wav')"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "zeO1xCqxUC6w"
},
"source": [
"## Entire audio inference"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "LjZBcsaTT7Mk"
},
"outputs": [],
"source": [
"wav = read_audio('en_example.wav', sampling_rate=SAMPLING_RATE)\n",
"# audio is being splitted into 31.25 ms long pieces\n",
"# so output length equals ceil(input_length * 31.25 / SAMPLING_RATE)\n",
"predicts = model.audio_forward(wav, sr=SAMPLING_RATE)"
]
},
{
"cell_type": "markdown",
"metadata": {
@@ -124,10 +138,10 @@
"source": [
"## using VADIterator class\n",
"\n",
"vad_iterator = VADIterator(model)\n",
"vad_iterator = VADIterator(model, sampling_rate=SAMPLING_RATE)\n",
"wav = read_audio(f'en_example.wav', sampling_rate=SAMPLING_RATE)\n",
"\n",
"window_size_samples = 1536 # number of samples in a single audio chunk\n",
"window_size_samples = 512 if SAMPLING_RATE == 16000 else 256\n",
"for i in range(0, len(wav), window_size_samples):\n",
" chunk = wav[i: i+ window_size_samples]\n",
" if len(chunk) < window_size_samples:\n",
@@ -150,7 +164,7 @@
"\n",
"wav = read_audio('en_example.wav', sampling_rate=SAMPLING_RATE)\n",
"speech_probs = []\n",
"window_size_samples = 1536\n",
"window_size_samples = 512 if SAMPLING_RATE == 16000 else 256\n",
"for i in range(0, len(wav), window_size_samples):\n",
" chunk = wav[i: i+ window_size_samples]\n",
" if len(chunk) < window_size_samples:\n",