ASR/create_freq_filter.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import torchaudio\n",
    "import os\n",
    "import torch\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f1f99e7deb0>]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "waveform, sample_rate = torchaudio.load('./test/audio/evala_00044.wav')\n",
    "fft = np.fft.fft(waveform.numpy()[0])\n",
    "amplitudes = np.abs(fft)\n",
    "\n",
    "ref_amplitude = 8423.331127865715\n",
    "amplitudes_dB = 20 * np.log10(amplitudes/ref_amplitude)\n",
    "freqs = np.fft.fftfreq(len(amplitudes))\n",
    "\n",
    "freqs_transformed = freqs * sample_rate\n",
    "\n",
    "# y scale must be log\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.xlim(0, 8000)\n",
    "plt.ylim(-90, 0)\n",
    "plt.plot(freqs_transformed, amplitudes_dB)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8423.331127865715"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "max(amplitudes) / ( 10 ** (-32.1/20) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f030446d2b0>]"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.percen"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 12000/12000 [02:01<00:00, 98.55it/s] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "min_center_freq: 5507.121951219512\n",
      "max_center_freq: 5507.121951219512\n",
      "min_bandwidth_fraction: 2.0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "test_set_folder = './test/audio/'\n",
    "\n",
    "test_set_audio_numbers = np.array([])\n",
    "min_freqs = np.array([])\n",
    "max_freqs = np.array([])\n",
    "\n",
    "for path in tqdm(os.listdir(test_set_folder)):\n",
    "    audio, sample_rate = librosa.load(os.path.join(test_set_folder, path))\n",
    "    spectrum = np.abs(librosa.stft(audio))\n",
    "    freqs = librosa.fft_frequencies(sr=sample_rate, n_fft=spectrum.shape[0])\n",
    "    test_set_audio_numbers = np.append(test_set_audio_numbers, int(path.split('_')[1][:-4]))\n",
    "    min_freqs = np.append(min_freqs, np.percentile(freqs))\n",
    "    max_freqs = np.append(max_freqs, np.max(freqs))\n",
    "\n",
    "centers = (min_freqs + max_freqs) / 2\n",
    "widths = max_freqs - min_freqs\n",
    "\n",
    "min_center_freq = np.min(centers)\n",
    "max_center_freq = np.max(centers)\n",
    "min_bandwidth_fraction = np.min(widths) / centers[np.argmin(widths)]\n",
    "\n",
    "print(f'min_center_freq: {min_center_freq}')\n",
    "print(f'max_center_freq: {max_center_freq}')\n",
    "print(f'min_bandwidth_fraction: {min_bandwidth_fraction}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# get the indices that will sort the test_set_audio_nums\n",
    "sorted_indices = np.argsort(test_set_audio_numbers)\n",
    "\n",
    "sorted_min_freqs = min_freqs[sorted_indices]\n",
    "sorted_max_freqs = max_freqs[sorted_indices]\n",
    "\n",
    "# plot the sorted_min_freqs and sorted_max_freqs as a line chart, with range(12000) as the x-axis\n",
    "plt.plot(range(12000), sorted_min_freqs, label='min_freqs')\n",
    "plt.plot(range(12000), sorted_max_freqs, label='max_freqs')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "til2023",
   "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": "3.9.16"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}