\n",
"\n",
"This tutorial is available as an IPython notebook at [malaya-speech/example/diarization-features](https://github.com/huseinzol05/malaya-speech/tree/master/example/diarization-features).\n",
" \n",
"
\n",
"\n",
"This is an application of malaya-speech Pipeline, read more about malaya-speech Pipeline at [malaya-speech/example/pipeline](https://github.com/huseinzol05/malaya-speech/tree/master/example/pipeline).\n",
" \n",
"
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### What is the different with Speaker Diarization\n",
"\n",
"Current speaker diarization, https://malaya-speech.readthedocs.io/en/latest/load-diarization.html\n",
"\n",
"Required a pipeline, VAD -> Group positive VADs -> Speaker models -> Clustering, and this pipeline required a really good VAD and Speaker models. What if we can directly cluster using STFT / Features and arange the timestamp.\n",
"\n",
"Inspired by [khursani8](https://github.com/khursani8),\n",
"\n",
"Wave -> STFT / Features -> Clustering -> arange timestamp.\n",
"\n",
"The features can be anything, such as,\n",
"\n",
"- MFCC\n",
"- Melspectrogram\n",
"- Conv"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/tensorflow_addons/utils/ensure_tf_install.py:67: UserWarning: Tensorflow Addons supports using Python ops for all Tensorflow versions above or equal to 2.3.0 and strictly below 2.5.0 (nightly versions are not supported). \n",
" The versions of TensorFlow you are currently using is 2.5.0 and is not supported. \n",
"Some things might work, some things might not.\n",
"If you were to encounter a bug, do not file an issue.\n",
"If you want to make sure you're using a tested and supported configuration, either change the TensorFlow version or the TensorFlow Addons's version. \n",
"You can find the compatibility matrix in TensorFlow Addon's readme:\n",
"https://github.com/tensorflow/addons\n",
" UserWarning,\n",
"/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/tensorflow_addons/utils/resource_loader.py:103: UserWarning: You are currently using TensorFlow 2.5.0 and trying to load a custom op (custom_ops/seq2seq/_beam_search_ops.so).\n",
"TensorFlow Addons has compiled its custom ops against TensorFlow 2.4.0, and there are no compatibility guarantees between the two versions. \n",
"This means that you might get segfaults when loading the custom op, or other kind of low-level errors.\n",
" If you do, do not file an issue on Github. This is a known limitation.\n",
"\n",
"It might help you to fallback to pure Python ops with TF_ADDONS_PY_OPS . To do that, see https://github.com/tensorflow/addons#gpucpu-custom-ops \n",
"\n",
"You can also change the TensorFlow version installed on your system. You would need a TensorFlow version equal to or above 2.4.0 and strictly below 2.5.0.\n",
" Note that nightly versions of TensorFlow, as well as non-pip TensorFlow like `conda install tensorflow` or compiled from source are not supported.\n",
"\n",
"The last solution is to find the TensorFlow Addons version that has custom ops compatible with the TensorFlow installed on your system. To do that, refer to the readme: https://github.com/tensorflow/addons\n",
" UserWarning,\n",
"/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/malaya_boilerplate-0.0.10-py3.7.egg/malaya_boilerplate/frozen_graph.py:28: UserWarning: Cannot import beam_search_ops from Tensorflow Addons, `deep_model` for stemmer will not available to use, make sure Tensorflow Addons version >= 0.12.0\n"
]
}
],
"source": [
"from malaya_speech import Pipeline\n",
"import malaya_speech\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Load audio sample"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(1634237, 16000)"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"y, sr = malaya_speech.load('speech/video/The-Singaporean-White-Boy.wav')\n",
"len(y), sr"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"# just going to take 60 seconds\n",
"y = y[:sr * 60]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This audio extracted from https://www.youtube.com/watch?v=HylaY5e1awo&t=2s"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Generate Log Melspectrogram\n",
"\n",
"You can use interface `malaya_speech.utils.featurization.STTFeaturizer`,\n",
"\n",
"```python\n",
"class STTFeaturizer:\n",
" def __init__(\n",
" self,\n",
" sample_rate=16000,\n",
" frame_ms=25,\n",
" stride_ms=10,\n",
" nfft=None,\n",
" num_feature_bins=80,\n",
" feature_type='log_mel_spectrogram',\n",
" preemphasis=0.97,\n",
" dither=1e-5,\n",
" normalize_signal=True,\n",
" normalize_feature=True,\n",
" norm_per_feature=True,\n",
" **kwargs,\n",
" ):\n",
" \"\"\"\n",
" sample_rate: int, optional (default=16000)\n",
" frame_ms: int, optional (default=25)\n",
" To calculate `frame_length` for librosa STFT, `frame_length = int(sample_rate * (frame_ms / 1000))`\n",
" stride_ms: int, optional (default=10)\n",
" To calculate `frame_step` for librosa STFT, `frame_step = int(sample_rate * (stride_ms / 1000))`\n",
" nfft: int, optional (default=None)\n",
" If None, will calculate by `math.ceil(math.log2((frame_ms / 1000) * sample_rate))`\n",
" num_feature_bins: int, optional (default=80)\n",
" Size of output features.\n",
" feature_type: str, optional (default='log_mel_spectrogram')\n",
" Features type, allowed values:\n",
"\n",
" * ``'spectrogram'`` - np.square(np.abs(librosa.core.stft))\n",
" * ``'mfcc'`` - librosa.feature.mfcc(np.square(np.abs(librosa.core.stft)))\n",
" * ``'log_mel_spectrogram'`` - log(mel(np.square(np.abs(librosa.core.stft))))\n",
" \"\"\"\n",
"```"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Spectral Clustering\n",
"\n",
"This is a Python re-implementation of the spectral clustering algorithm in the paper [Speaker Diarization with LSTM](https://google.github.io/speaker-id/publications/LstmDiarization/).\n",
"\n",
"So, make sure you already install [spectralcluster](https://pypi.org/project/spectralcluster/),\n",
"\n",
"```bash\n",
"pip install spectralcluster\n",
"```"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {},
"outputs": [],
"source": [
"from spectralcluster import SpectralClusterer\n",
"\n",
"clusterer = SpectralClusterer(\n",
" min_clusters=1,\n",
" max_clusters=100,\n",
" p_percentile=0.95,\n",
" gaussian_blur_sigma=30.0,\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Clustering on log MelSpectrogram"
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"CPU times: user 110 ms, sys: 31.2 ms, total: 141 ms\n",
"Wall time: 111 ms\n"
]
},
{
"data": {
"text/plain": [
"(2001, 80)"
]
},
"execution_count": 43,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"%%time\n",
"\n",
"featurizer = malaya_speech.featurization.STTFeaturizer(feature_type = 'log_mel_spectrogram',\n",
" frame_ms = 50, stride_ms = 30)\n",
"features = featurizer(y)\n",
"features.shape"
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {},
"outputs": [],
"source": [
"from malaya_speech.utils.dist import l2_normalize"
]
},
{
"cell_type": "code",
"execution_count": 45,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"CPU times: user 15.3 s, sys: 1.56 s, total: 16.8 s\n",
"Wall time: 5.2 s\n"
]
}
],
"source": [
"%%time\n",
"\n",
"cluster_labels = clusterer.predict(l2_normalize(features))\n",
"frames = malaya_speech.arange.arange_frames(features, y, sr)\n",
"results = []\n",
"for no, result in enumerate(cluster_labels):\n",
" results.append((frames[no], result))\n",
"grouped = malaya_speech.group.group_frames(results)"
]
},
{
"cell_type": "code",
"execution_count": 46,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[(, 1),\n",
" (, 2),\n",
" (, 1),\n",
" (, 2),\n",
" (, 1),\n",
" (, 2),\n",
" (, 0),\n",
" (, 1),\n",
" (, 0),\n",
" (, 1),\n",
" (, 0),\n",
" (, 2),\n",
" (, 0),\n",
" (, 1)]"
]
},
"execution_count": 46,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"grouped"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Clustering on TRILL\n",
"\n",
"The TRILL model presented in \"Towards Learning a Universal Non-Semantic Representation of Speech\". It exceeds state-of-the-art performance on a number of transfer learning tasks drawn from the non-semantic speech domain (speech emotion recognition, language identification, etc). It is trained on publicly-available AudioSet, https://tfhub.dev/google/nonsemantic-speech-benchmark/trill/3"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"import tensorflow_hub as hub\n",
"module = hub.load('https://tfhub.dev/google/nonsemantic-speech-benchmark/trill/3')"
]
},
{
"cell_type": "code",
"execution_count": 47,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"2000"
]
},
"execution_count": 47,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# i think 60ms pretty ok\n",
"frames = malaya_speech.generator.frames(y, frame_duration_ms = 30)\n",
"len(frames)"
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|██████████| 2000/2000 [01:02<00:00, 32.23it/s]\n"
]
}
],
"source": [
"from tqdm import tqdm\n",
"\n",
"arrays = [f.array for f in frames]\n",
"embeddings = []\n",
"for i in tqdm(range(len(arrays))):\n",
" e = module(arrays[i], sample_rate=16000)['embedding']\n",
" embeddings.append(e)"
]
},
{
"cell_type": "code",
"execution_count": 49,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(2000, 512)"
]
},
"execution_count": 49,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"concat = np.concatenate(embeddings, axis = 0)\n",
"concat.shape"
]
},
{
"cell_type": "code",
"execution_count": 65,
"metadata": {},
"outputs": [],
"source": [
"clusterer = SpectralClusterer(\n",
" min_clusters=1,\n",
" max_clusters=100,\n",
" p_percentile=0.95,\n",
" gaussian_blur_sigma=1.0,\n",
" thresholding_soft_multiplier = 1.0,\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 66,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"CPU times: user 16.1 s, sys: 1.43 s, total: 17.5 s\n",
"Wall time: 4.28 s\n"
]
},
{
"data": {
"text/plain": [
"[(, 0)]"
]
},
"execution_count": 66,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"%%time\n",
"\n",
"cluster_labels = clusterer.predict(l2_normalize(concat))\n",
"frames = malaya_speech.arange.arange_frames(concat, y, sr)\n",
"results_trill = []\n",
"for no, result in enumerate(cluster_labels):\n",
" results_trill.append((frames[no], result))\n",
"grouped_trill = malaya_speech.group.group_frames(results_trill)\n",
"grouped_trill"
]
},
{
"cell_type": "code",
"execution_count": 67,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/Users/huseinzolkepli/Documents/malaya-speech/malaya_speech/extra/visualization.py:168: RuntimeWarning: invalid value encountered in true_divide\n",
" std = (a - np.min(a)) / (np.max(a) - np.min(a))\n"
]
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"nrows = 3\n",
"fig, ax = plt.subplots(nrows = nrows, ncols = 1)\n",
"fig.set_figwidth(20)\n",
"fig.set_figheight(nrows * 3)\n",
"min_timestamp = min([i[0].timestamp for i in grouped])\n",
"max_timestamp = max([i[0].timestamp + i[0].duration for i in grouped])\n",
"ax[0].set_xlim((min_timestamp, max_timestamp))\n",
"ax[0].plot([i / sr for i in range(len(y))], y)\n",
"malaya_speech.extra.visualization.plot_classification(grouped, \n",
" 'diarization using spectral cluster', ax = ax[1],\n",
" x_text = 0.01)\n",
"malaya_speech.extra.visualization.plot_classification(grouped_trill, \n",
" 'diarization using spectral cluster TRILL', ax = ax[2],\n",
" x_text = 0.01)\n",
"fig.tight_layout()\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 68,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 68,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import IPython.display as ipd\n",
"\n",
"ipd.Audio(grouped[0][0].array, rate = sr)"
]
},
{
"cell_type": "code",
"execution_count": 69,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 69,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"ipd.Audio(grouped[1][0].array, rate = sr)"
]
},
{
"cell_type": "code",
"execution_count": 70,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 70,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"ipd.Audio(grouped[2][0].array, rate = sr)"
]
},
{
"cell_type": "code",
"execution_count": 71,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 71,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"ipd.Audio(grouped[3][0].array, rate = sr)"
]
}
],
"metadata": {
"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": "3.7.7"
}
},
"nbformat": 4,
"nbformat_minor": 4
}