week07/SemRealTimeTOSOLVE.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rRgUY8rKrJ06"
      },
      "source": [
        "## Causal vs non-Causal"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "q7fJ54YMGhN2",
        "outputId": "060b989d-f94b-42d2-a581-073e0c9e111f"
      },
      "outputs": [],
      "source": [
        "!pip install -U denoiser"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "njuW9OWX5Po_"
      },
      "source": [
        "![image.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "7y6sDj2rmvEn"
      },
      "outputs": [],
      "source": [
        "from IPython import display as disp\n",
        "import torch\n",
        "import torchaudio\n",
        "from denoiser import pretrained\n",
        "from denoiser.dsp import convert_audio\n",
        "from scipy.signal import wiener, convolve\n",
        "from scipy.special import sinc\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import scipy\n",
        "import matplotlib"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_ifyFFLYmySG",
        "outputId": "2a4b2a57-1e66-46c5-d878-09f837342432"
      },
      "outputs": [],
      "source": [
        "# Downloading a test file\n",
        "!wget https://facebookresearch.github.io/denoiser/audio/noisy/alex_noisy.mp3"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 87
        },
        "id": "jIClnmB-m0m9",
        "outputId": "1bf20287-c47a-402f-96cb-e91da2dc1038"
      },
      "outputs": [],
      "source": [
        "model = pretrained.dns64()\n",
        "wav, sr = torchaudio.load('./alex_noisy.mp3',format=\"mp3\")\n",
        "wav = convert_audio(wav, sr, model.sample_rate, model.chin)\n",
        "with torch.no_grad():\n",
        "    causal_denoised = model(wav[None])[0]\n",
        "disp.display(disp.Audio(wav.numpy(), rate=model.sample_rate))\n",
        "disp.display(disp.Audio(causal_denoised.numpy(), rate=model.sample_rate))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 87
        },
        "id": "RMPqOACbnaxc",
        "outputId": "5ce2855c-0a9a-43cb-c6c5-4566d49f47b3"
      },
      "outputs": [],
      "source": [
        "wiener_denoised = wiener(wav.squeeze().numpy(), 7)\n",
        "disp.display(disp.Audio(wav.numpy(), rate=sr))\n",
        "disp.display(disp.Audio(wiener_denoised, rate=sr))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 932
        },
        "id": "pmN_rMrSIG8X",
        "outputId": "fcb5d1cd-c5a3-460b-e07a-0326378ee17c"
      },
      "outputs": [],
      "source": [
        "n_fft = 1024\n",
        "window_size = n_fft\n",
        "hop_size = 256\n",
        "window = torch.hann_window(window_size)\n",
        "\n",
        "def plot_specs(wav1, wav2):\n",
        "    freqs1, ts1, spec1 = scipy.signal.stft(wav1, sr, nperseg=n_fft)\n",
        "    freqs2, ts2, spec2 = scipy.signal.stft(wav2, sr, nperseg=n_fft)\n",
        "    norm_factor = np.amax(np.abs(spec1))/np.amax(np.abs(spec2))\n",
        "    spec2 = spec2 * norm_factor\n",
        "\n",
        "    fig, (ax11,ax21) = plt.subplots(2,1,figsize=(13,11))\n",
        "\n",
        "    ax11.pcolormesh(ts1, freqs1, np.log(np.abs(spec1) + 1e-6),  shading='gouraud', cmap=matplotlib.colormaps['gist_heat'])\n",
        "    ax11.set_ylabel('Frequency (Hz)')\n",
        "    ax11.set_xlabel('Time')\n",
        "\n",
        "    im2=ax21.pcolormesh(ts2, freqs2, np.log(np.abs(spec2) + 1e-6),  shading='gouraud', cmap=matplotlib.colormaps['gist_heat'])\n",
        "    ax21.set_ylabel('Frequency [Hz]')\n",
        "    ax21.set_xlabel('Time')\n",
        "\n",
        "    plt.show()\n",
        "\n",
        "def visualize_audio(wav, sr):\n",
        "    # Average all channels\n",
        "    if wav.dim() == 2:\n",
        "        # Any to mono audio convertion\n",
        "        wav = wav.mean(dim=0)\n",
        "\n",
        "    plt.figure(figsize=(20, 5))\n",
        "    plt.plot(wav, alpha=.7, c='green')\n",
        "    plt.grid()\n",
        "    plt.xlabel('Time', size=20)\n",
        "    plt.ylabel('Amplitude', size=20)\n",
        "    plt.show()\n",
        "\n",
        "    disp.display(disp.Audio(wav, rate=sr))\n",
        "\n",
        "plot_specs(wav.squeeze().numpy(), wiener_denoised.squeeze())\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zolB7Qp-nLIv"
      },
      "source": [
        "## Sinc filter\n",
        "\n",
        "With parameter $B$  is defined as\n",
        "\n",
        "$$\n",
        "    h(t) = 2B sinc(2Bt)\n",
        "$$\n",
        "\n",
        "where $sinc$ is the normalized sinc function\n",
        "\n",
        "$$\n",
        "    sinc(x) = {sin(\\pi t) \\over \\pi t}\n",
        "$$\n",
        "\n",
        "When FT is taken it turns out that it is a rectangular frequency filter. Thus, it removes all frequencies above some level $B$:\n",
        "$$\n",
        "H(f) = \\begin{cases}\n",
        "            0, & |f| > B\\\\\n",
        "            {1 \\over 2}, & |f| = B\\\\\n",
        "            1, & |f| < B\n",
        "        \\end{cases}\n",
        "$$"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "One of the constructions of rectangular low- and hi-pass filters comes from the usage of the sinc function."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 599
        },
        "id": "1ygu2x50oUK3",
        "outputId": "93baf6e5-f4fb-4b75-d007-fd3ec63986bd"
      },
      "outputs": [],
      "source": [
        "fc = 0.2\n",
        "fc1 = 0.1\n",
        "N = wav.shape[1]\n",
        "sinc_filter = 2 * fc * sinc(2 * fc * (np.arange(N) - (N - 1) / 2.))\n",
        "sinc_filter1 = 2 * fc1 * sinc(2 * fc1 * (np.arange(N) - (N - 1) / 2.))\n",
        "\n",
        "plt.figure(figsize=(4, 4))\n",
        "plt.plot((np.arange(N) - (N - 1) / 2.),sinc_filter)\n",
        "plt.grid(True)\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 932
        },
        "id": "Hg75sYBDrjvh",
        "outputId": "7110255b-65b9-4f33-9d1f-5ad3dac66c28"
      },
      "outputs": [],
      "source": [
        "filtered_wav = convolve(wav.squeeze().numpy(), sinc_filter - sinc_filter1, mode=\"same\")\n",
        "\n",
        "plot_specs(wav.squeeze().numpy(), filtered_wav)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 599
        },
        "id": "7ZnJO6FzQY4G",
        "outputId": "38f83d25-8965-45fe-b993-a3c1ff003a39"
      },
      "outputs": [],
      "source": [
        "fc = 0.2\n",
        "N = 201\n",
        "sinc_filter_small = sinc(2 * fc * (np.arange(N) - (N - 1) / 2.))\n",
        "\n",
        "plt.figure(figsize=(7, 7))\n",
        "plt.plot(sinc_filter_small)\n",
        "plt.grid(True)\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 932
        },
        "id": "G2vKcsN9KJI0",
        "outputId": "b10a05b8-5aab-418e-b94d-32cc1ac09d98"
      },
      "outputs": [],
      "source": [
        "filtered_wav = convolve(wav.squeeze().numpy(), sinc_filter_small, mode=\"same\")\n",
        "\n",
        "plot_specs(wav.squeeze().numpy(), filtered_wav)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "thcKRHKlRMb8",
        "outputId": "d467e85a-8bb5-4ac8-f125-1a7ce5019f66"
      },
      "outputs": [],
      "source": [
        "w = np.blackman(N)\n",
        "sinc_filter_small = sinc_filter_small * w\n",
        "sinc_filter_small /= sinc_filter_small.sum()\n",
        "\n",
        "plt.figure(figsize=(7, 7))\n",
        "plt.plot(sinc_filter_small)\n",
        "plt.grid(True)\n",
        "plt.show()\n",
        "\n",
        "filtered_wav = convolve(wav.squeeze().numpy(), sinc_filter_small, mode=\"same\")\n",
        "\n",
        "\n",
        "plot_specs(wav.squeeze().numpy(), filtered_wav)\n",
        "# plot_spec(torch.from_numpy(filtered_wav))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 864
        },
        "id": "Ag2rL7DC4bst",
        "outputId": "fbfaf1b9-6cca-4667-8065-f097f63f9ced"
      },
      "outputs": [],
      "source": [
        "visualize_audio(torch.from_numpy(filtered_wav), sr)\n",
        "visualize_audio(wav, sr)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 87
        },
        "id": "pwhiBbE6PPoQ",
        "outputId": "1aec7736-0984-4ede-cd7b-eeb73800db77"
      },
      "outputs": [],
      "source": [
        "model_nc = pretrained.valentini_nc()\n",
        "\n",
        "with torch.no_grad():\n",
        "    nc_denoised = model_nc(wav[None])[0]\n",
        "disp.display(disp.Audio(wav.numpy(), rate=model.sample_rate))\n",
        "disp.display(disp.Audio(nc_denoised.numpy(), rate=model.sample_rate))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 932
        },
        "id": "AGOQ2lxRYgWM",
        "outputId": "d7588a19-3e9e-4013-99d4-6eb10643d6cb"
      },
      "outputs": [],
      "source": [
        "plot_specs(causal_denoised.squeeze().numpy(), nc_denoised.squeeze().numpy())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 87
        },
        "id": "lGGfqIXnbUB2",
        "outputId": "12c8df18-3171-4754-8385-9dba4b191968"
      },
      "outputs": [],
      "source": [
        "frame_size = 16 * 40 # 40ms\n",
        "n_chunks = wav.shape[1] // frame_size\n",
        "\n",
        "chunked_wav = torch.stack(torch.chunk(wav.squeeze(), n_chunks)[:-1], dim=0)\n",
        "print(chunked_wav.shape)\n",
        "with torch.no_grad():\n",
        "    nc_denoised_chunked = model_nc(chunked_wav)\n",
        "    causal_denoised_chunked = model(chunked_wav)\n",
        "disp.display(disp.Audio(causal_denoised_chunked.reshape(-1).numpy(), rate=model.sample_rate))\n",
        "disp.display(disp.Audio(nc_denoised_chunked.reshape(-1).numpy(), rate=model.sample_rate))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 442
        },
        "id": "rCBkyBhbKgf3",
        "outputId": "ce5e0b41-b74c-44ba-c036-f1fcc4e898c3"
      },
      "outputs": [],
      "source": [
        "import librosa\n",
        "\n",
        "def overlap_add(model, wav: np.ndarray, chunk_size):\n",
        "    stride = chunk_size // 2\n",
        "    n = len(wav) - chunk_size\n",
        "    n_chunks = n // stride + 1 + (n % stride > 0)\n",
        "    n_pad = stride - n % stride\n",
        "\n",
        "    wav_padded = np.pad(wav, (0, n_pad), 'constant', constant_values=(0, 1e-5))\n",
        "\n",
        "    chunks = torch.from_numpy(librosa.util.frame(wav_padded, frame_length=chunk_size, hop_length=stride)).permute(1, 0)\n",
        "\n",
        "    with torch.no_grad():\n",
        "        chunks = model(chunks)[:, 0, :]\n",
        "\n",
        "    n_chunks = chunks.shape[0]\n",
        "    first = chunks[0, 0:stride]\n",
        "    last = chunks[-1, stride:]\n",
        "\n",
        "    overlapped_chunks = chunks.reshape(-1)[stride:chunk_size * (n_chunks - 1) + stride].reshape(-1, chunk_size)\n",
        "    window_vec = torch.sin(torch.arange(stride) / (stride - 1)) * (np.pi / 2)\n",
        "    kernel_matrix = torch.zeros(2*stride, stride, dtype=overlapped_chunks.dtype)\n",
        "    kernel_matrix[:stride, :] = torch.diag(1 - window_vec)\n",
        "    kernel_matrix[stride:, :] = torch.diag(window_vec)\n",
        "    sum_overlapped_chunks = (overlapped_chunks @ kernel_matrix).reshape(-1)\n",
        "\n",
        "    overlap_added_wav = torch.cat((first, sum_overlapped_chunks, last), dim=-1)\n",
        "    return overlap_added_wav\n",
        "\n",
        "overadd_wav = overlap_add(model, wav.squeeze().numpy(), chunk_size=650)\n",
        "visualize_audio(overadd_wav, sr)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 440
        },
        "id": "oRJgYuNJN-5c",
        "outputId": "48a2a831-6040-431b-d639-b63487c721f0"
      },
      "outputs": [],
      "source": [
        "overadd_wav = overlap_add(model_nc, wav.squeeze().numpy(), chunk_size=650)\n",
        "visualize_audio(overadd_wav, sr)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 440
        },
        "id": "1zNlYct2jefC",
        "outputId": "4e490f5d-7351-4619-bbfb-6f992788e43f"
      },
      "outputs": [],
      "source": [
        "from denoiser.demucs import DemucsStreamer\n",
        "\n",
        "streamer = DemucsStreamer(model)\n",
        "with torch.no_grad():\n",
        "    estimate = torch.cat([\n",
        "        streamer.feed(wav),\n",
        "        streamer.flush()], dim=1)[None]\n",
        "\n",
        "visualize_audio(estimate.squeeze(), sr)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LeBnXjc68Xgn"
      },
      "source": [
        "In original demucs implementation there are 3 main techniques to enhance quality on evaluation of real-time model:\n",
        "- Normalize audio with variance estimated during evaluation;\n",
        "\n",
        "- Padding with history before upsampling and downsampling;\n",
        "\n",
        "- Return extra strides from model inference."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "PjDmztHBu24O"
      },
      "outputs": [],
      "source": [
        "import math\n",
        "from denoiser.resample import upsample2, downsample2\n",
        "\n",
        "def demucs_process(model, wav):\n",
        "    streamer = DemucsStreamer(model)\n",
        "    resample_buffer = streamer.resample_buffer\n",
        "    stride = streamer.stride\n",
        "    resample = model.resample\n",
        "    n_frames = 0\n",
        "    total_variance = 0\n",
        "\n",
        "    resample_in = torch.zeros(1, resample_buffer)\n",
        "    resample_out = torch.zeros(1, resample_buffer)\n",
        "\n",
        "    stream = wav\n",
        "    outs = []\n",
        "    while stream.shape[1] >= streamer.total_length:\n",
        "        n_frames += 1\n",
        "        frame = stream[:, :streamer.total_length]\n",
        "        if model.normalize:\n",
        "            mono = frame.mean(0)\n",
        "            variance = (mono**2).mean()\n",
        "            total_variance = variance / n_frames + (1 - 1 / n_frames) * total_variance # update variance\n",
        "            frame = frame / (model.floor + math.sqrt(total_variance)) # norm audio by variance\n",
        "        padded_frame = torch.cat([resample_in, frame], dim=-1)\n",
        "        resample_in[:] = frame[:, stride - resample_buffer:stride]\n",
        "        frame = padded_frame\n",
        "\n",
        "        if resample == 4:\n",
        "            frame = upsample2(upsample2(frame))\n",
        "        elif resample == 2:\n",
        "            frame = upsample2(frame)\n",
        "        frame = frame[:, resample * resample_buffer:]  # remove pre sampling buffer\n",
        "        frame = frame[:, :resample * streamer.frame_length]  # remove extra samples after window\n",
        "\n",
        "        out, extra = streamer._separate_frame(frame)\n",
        "        padded_out = torch.cat([resample_out, out, extra], 1)\n",
        "        resample_out[:] = out[:, -resample_buffer:]\n",
        "        if resample == 4:\n",
        "            out = downsample2(downsample2(padded_out))\n",
        "        elif resample == 2:\n",
        "            out = downsample2(padded_out)\n",
        "        else:\n",
        "            out = padded_out\n",
        "\n",
        "        out = out[:, resample_buffer // resample:]\n",
        "        out = out[:, :stride]\n",
        "\n",
        "        if model.normalize:\n",
        "            out *= math.sqrt(total_variance)\n",
        "\n",
        "        outs.append(out)\n",
        "        stream = stream[:, stride:]\n",
        "\n",
        "    out = torch.cat(outs, 1)\n",
        "\n",
        "    return out"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "sn4rmBqQ3Hlq"
      },
      "outputs": [],
      "source": [
        "with torch.no_grad():\n",
        "    estimate = demucs_process(model, wav)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 440
        },
        "id": "FeI5XnUe3MVr",
        "outputId": "6e9ac147-7b52-4b48-9293-d25ea5241340"
      },
      "outputs": [],
      "source": [
        "visualize_audio(estimate.squeeze(), sr)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Complexity estimation\n",
        "\n",
        "There are certain metrics to compute how suitable you are for realtime setting: they include both model complexity and performance metrics.\n",
        "\n",
        "* RTF(realtime factor): time of processing of 1 chunk of audio divided by audio length (in seconds)\n",
        "* Processing time (per second of audio)\n",
        "* FLOps or MACs: number of arithmetic operations needed to pproduce a result (with input of 1 sec audio)\n",
        "* Model complexity: number of parameters and/or size in bytes"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "vQbiAXI23XaI"
      },
      "outputs": [],
      "source": [
        "from thop import profile\n",
        "\n",
        "def compute_gflops_and_model_size(model,inputSample):\n",
        "    macs, params = profile(model, inputs=(inputSample,), verbose=False)\n",
        "    #GFlops = macs * 2.0 / pow(10, 9)#a rough estimate\n",
        "    GMacs = macs / 1e9\n",
        "    params_M = params/1e6\n",
        "    return params_M, GMacs"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "print(\"SR\", sr)\n",
        "wav1Sec = wav[...,:sr]\n",
        "print(wav1Sec)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "compute_gflops_and_model_size(model_nc,wav1Sec)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "compute_gflops_and_model_size(model,wav1Sec)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Demucs also has a short way to compute delay and RTF on your device:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "!python -m denoiser.demucs --hidden=48 --num_threads=4"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "It is given in detail [in the repo](https://github.com/facebookresearch/denoiser/blob/main/denoiser/demucs.py)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "studyEnv",
      "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": "3.10.14"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}