{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "%matplotlib inline"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "\n# Add Noise to Speech at Specific SNR Levels\n\n\nAdd noise to speech at specific signal-to-noise ratio levels.\n\nTo see how soundpy implements this, see `soundpy.dsp.add_backgroundsound`.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Let's import soundpy, and ipd for playing audio data\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "import soundpy as sp\nimport IPython.display as ipd"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Define the speech and noise data samples\n^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "I will use speech and noise data from the soundpy repo.\n\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Designate path relevant for accessing audiodata\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "sp_dir = '../../../'"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Speech sample:\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "speech_sample = '{}audiodata/python.wav'.format(sp_dir)\nspeech_sample = sp.utils.string2pathlib(speech_sample)\n# as pathlib object, can do the following: \nword = speech_sample.stem\nword"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Noise sample:\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "noise_sample = '{}audiodata/background_samples/cafe.wav'.format(sp_dir)\nnoise_sample = sp.utils.string2pathlib(noise_sample)\n# as pathlib object, can do the following: \nnoise = noise_sample.stem\nnoise"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Hear Clean Speech\n~~~~~~~~~~~~~~~~~\nI'm using a higher sample rate here as calculating SNR \nperforms best upwards of 44100 Hz.\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "sr = 44100\ns, sr = sp.loadsound(speech_sample, sr = sr)\nipd.Audio(s,rate=sr)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Hear Noise\n~~~~~~~~~~\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "n, sr = sp.loadsound(noise_sample, sr = sr)\nipd.Audio(n,rate=sr)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Hear Signal-to-Noise Ratio 20\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "noisyspeech_20snr, snr20 = sp.dsp.add_backgroundsound(\n    speech_sample,\n    noise_sample,\n    sr = sr,\n    snr = 20)\nipd.Audio(noisyspeech_20snr,rate=sr)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "`snr20` is simply the measured SNR post adjustment fo the noise signal.\nThis is useful to check that the indicated snr is at least close\nto the resulting snr.\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "snr20"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Hear Signal-to-Noise Ratio 5\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "noisyspeech_5snr, snr5 = sp.dsp.add_backgroundsound(\n    speech_sample,\n    noise_sample,\n    sr = sr,\n    snr = 5)\nipd.Audio(noisyspeech_5snr,rate=sr)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "snr5"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Visualize the Audio Samples\n^^^^^^^^^^^^^^^^^^^^^^^^^^^\n\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "See Clean Speech (raw signal)\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "sp.plotsound(speech_sample, feature_type='signal', \n               sr = sr, title = 'Speech: ' + word.upper(),\n               subprocess=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "See Clean Speech (stft)\n~~~~~~~~~~~~~~~~~~~~~~~\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "sp.plotsound(speech_sample, feature_type='stft', \n               sr = sr, title = 'Speech: ' + word.upper(), \n               subprocess=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "~~~~~~~~~~~~~~~~~~~~~~\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "sp.plotsound(noise_sample, feature_type='signal',\n               title = 'Noise: ' + noise.upper(), subprocess=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "~~~~~~~~~~~~~~~~\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "sp.plotsound(noise_sample, feature_type='stft',\n               title = 'Noise: ' + noise.upper(), subprocess=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "See Noisy Speech: SNR 20 (raw signal)\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "sp.plotsound(noisyspeech_20snr, sr = sr, feature_type = 'signal',\n               title = '\"{}\" with {} noise at SNR 20'.format(word.upper(), noise.upper()),\n               subprocess=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "See Noisy Speech: SNR 20 (stft)\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "sp.plotsound(noisyspeech_20snr, sr = sr, feature_type = 'stft',\n               title = '\"{}\" with {} noise at SNR 20'.format(word.upper(), noise.upper()),\n               subprocess=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "See Noisy Speech: SNR 5 (raw signal)\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "sp.plotsound(noisyspeech_5snr, sr = sr, feature_type = 'signal',\n               title = '\"{}\" with {} noise at SNR 5'.format(word.upper(), noise.upper()),\n               subprocess=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "See Noisy Speech: SNR 5 (stft)\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "sp.plotsound(noisyspeech_20snr, sr = sr, feature_type = 'stft',\n               title = '\"{}\" with {} noise at SNR 5'.format(word.upper(), noise.upper()),\n               subprocess=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Make Combined Sound Longer\n^^^^^^^^^^^^^^^^^^^^^^^^^^\n\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Pad Speech and Set Total Length \n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "noisyspeech_20snr, snr20 = sp.dsp.add_backgroundsound(\n    speech_sample,\n    noise_sample,\n    sr = sr,\n    snr = 20,\n    pad_mainsound_sec = 1,\n    total_len_sec = 4)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "ipd.Audio(noisyspeech_20snr,rate=sr)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "sp.plotsound(noisyspeech_20snr, sr = sr, feature_type = 'signal',\n               title = '\"{}\" with {} noise at SNR 20'.format(word.upper(), noise.upper()),\n               subprocess=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Make Combined Sound Shorter\n^^^^^^^^^^^^^^^^^^^^^^^^^^^\n\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Set Total Length\n~~~~~~~~~~~~~~~~\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "noisyspeech_20snr, snr20 = sp.dsp.add_backgroundsound(\n    speech_sample,\n    noise_sample,\n    sr = sr,\n    snr = 20,\n    total_len_sec = 0.75)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "ipd.Audio(noisyspeech_20snr,rate=sr)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "sp.plotsound(noisyspeech_20snr, sr = sr, feature_type = 'signal',\n               title = '\"{}\" with {} noise at SNR 20'.format(word.upper(), noise.upper()),\n               subprocess=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Wrap the Background Sound\n^^^^^^^^^^^^^^^^^^^^^^^^^\n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "noisyspeech_20snr, snr20 = sp.dsp.add_backgroundsound(\n    speech_sample,\n    noise_sample,\n    sr = sr,\n    snr = 20,\n    wrap = True,\n    pad_mainsound_sec = 2,\n    total_len_sec = 5)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "ipd.Audio(noisyspeech_20snr,rate=sr)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "sp.plotsound(noisyspeech_20snr, sr = sr, feature_type = 'signal',\n               title = '\"{}\" with {} noise at SNR 20'.format(word.upper(), noise.upper()),\n               subprocess=True)"
      ]
    }
  ],
  "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": "3.8.2"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}