\n",
"Following Section 8.2 and Section 8.2.3.3 of [Müller, FMP, Springer 2015], we cover in this notebook two related problems referred to as melody extraction and melody separation. The overall approach was suggested by Salamon and Gómez:\n",
" \n",
"
\n",
"
\n",
"Justin Salamon and Emilia Gómez: Melody Extraction from Polyphonic Music Signals using Pitch Contour Characteristics. IEEE Transactions on Audio, Speech, and Language Processing, 20 (2012), pp. 1759–1770.\n",
" \n",
" Bibtex \n",
"
\n",
"
\n",
" \n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Melody\n",
"\n",
"When asked to describe a specific song, we are often able to sing or hum the main melody. In general terms, a **melody** may be defined as a linear succession of musical tones expressing a particular musical idea. Because of the special arrangement of tones, a melody is perceived as a coherent entity, which gets stuck in a listener's head as the most memorable element of a song. As the original Greek term **melōidía** (meaning \"singing\" or \"chanting\") implies, a melody is often performed by a human voice. Oftentimes, the melody constitutes the leading element in a composition, appearing in the foreground, while the accompaniment is in the background. Sometimes melody and accompaniment may even be played on a single instrument such as a guitar or a piano. In any case, the melody typically stands out in one way or another. For example, the melody often comprises the higher notes in a musical composition, while the accompaniment consists of the lower notes. Or the melody is played by some instrument with a characteristic timbre . In some performances, the notes of a melody may feature easily discernible time–frequency patterns such as **vibrato**, **tremolo**, or **glissando** (i.e., a continuous glide from one pitch to another). As with many other concepts in music processing, the notion of melody remains rather vague. In music processing, one often considers a restricted scenario. In the [FMP notebook on F0 tracking](../C8/C8S2_FundFreqTracking.html), for example, we make the following simplifying assumptions:\n",
"\n",
"* First, we consider the scenario where the music is given in the form of an audio recording (and not as a symbolic music representation).\n",
"* Second, rather than estimating a sequence of notes, we represent the melody by a sequence of F0-values that correspond to the notes' pitches.\n",
"* Third, we restrict ourselves to music where the melody is predominant (e.g., being performed by a lead singer or a lead instrument). \n",
"* Forth, we assume that there is only one melody line at a time, which can be associated to a single sound source."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Melody Extraction\n",
"\n",
"Based on these assumptions, the **melody extraction problem** can be regarded as the following signal processing task: Given a recording, the objective is to automatically estimate the sequence of **predominant F0-values** that correspond to the notes played by the lead voice or instrument. This is exactly the problem we discussed in the [FMP notebook on F0 tracking](../C8/C8S2_FundFreqTracking.html). The basic procedure can be summarized as follows:\n",
"\n",
"* First, compute a [salience representation](../C8/C8S2_SalienceRepresentation.html).\n",
"* Then, derive the [trajectory](../C8/C8S2_FundFreqTracking.html) of predominant F0-values using suitable smoothness or score-informed constraints. \n",
"* Also, as suggested by Salamon and Gómez, one may introduce other musical constraints, e.g., based on pitch contours. \n",
"\n",
"The melody extraction problem is challenging even in the restricted case that the simplifying assumptions are valid. \n",
"\n",
"* In music with many instruments playing simultaneously, it is hard to attribute specific time–frequency patterns to notes of individual instruments.\n",
"* Also, the presence of resonance and reverberation effects further increases the overlap of different sound sources. \n",
"* Furthermore, even after a successful estimation of fundamental frequencies, one still has to determine which of the F0-values belong to the predominant melody and which are part of the accompaniment.\n",
"\n",
"Continuing our [Freischütz example](../C8/C8S2_SalienceRepresentation.html), the following figure shows a spectrogram representation as well as the F0-trajectory of the melody performed by a soprano singer. Note the following: \n",
"\n",
"* The singer is not always active. This is also indicated by the activity pattern plotted below the spectrogram.\n",
"\n",
"* At some time points, the spectral bin with the highest magnitude may belong to the accompaniment. This is one reason for errors in melody estimation. \n",
"\n",
"* Some of the higher harmonics may contain more energy than the fundamental frequency, which may lead to **octave errors** in melody estimation. "
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"execution": {
"iopub.execute_input": "2024-02-15T08:43:47.517091Z",
"iopub.status.busy": "2024-02-15T08:43:47.516758Z",
"iopub.status.idle": "2024-02-15T08:43:50.173224Z",
"shell.execute_reply": "2024-02-15T08:43:50.172551Z"
}
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"\n",
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 1,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import numpy as np\n",
"import os, sys, librosa\n",
"from scipy import signal\n",
"from scipy import ndimage\n",
"from scipy.interpolate import interp1d\n",
"import matplotlib\n",
"from matplotlib import pyplot as plt\n",
"from matplotlib.colors import LinearSegmentedColormap, BoundaryNorm\n",
"from matplotlib import gridspec\n",
"import IPython.display as ipd\n",
"from numba import jit\n",
"import pandas as pd\n",
"from collections import OrderedDict\n",
"\n",
"sys.path.append('..')\n",
"import libfmp.b\n",
"import libfmp.c3\n",
"import libfmp.c8\n",
"\n",
"%matplotlib inline\n",
"\n",
"def plot_STFT_F0_activity(Y, traj, figsize=(7,4), ylim = [0, 4000], cmap='gray_r'):\n",
" fig, ax = plt.subplots(2, 2, gridspec_kw={'width_ratios': [10, 0.2],\n",
" 'height_ratios': [10, 1]}, figsize= figsize) \n",
" \n",
" fig_im, ax_im, im = libfmp.b.plot_matrix(Y, Fs=Fs/H, Fs_F=N/Fs, \n",
" ax=[ax[0,0], ax[0,1]], title='', cmap=cmap);\n",
" ax[0,0].set_ylim(ylim)\n",
" traj_plot = traj[traj[:, 1] > 0, :]\n",
" ax[0,0].plot(traj_plot[:, 0], traj_plot[:, 1], color='r', markersize=4, marker='.', linestyle='');\n",
" \n",
" # F0 activity\n",
" activity = np.array(traj[:, 1] > 0).reshape(1, -1)\n",
" im = ax[1,0].imshow(activity, aspect='auto', cmap='bwr', clim=[-1, 1])\n",
" ax[1,0].set_xticks([])\n",
" ax[1,0].set_yticks([])\n",
" ax[1,1].axis('off')\n",
" plt.tight_layout()\n",
" return fig, ax\n",
"\n",
"# Load audio\n",
"fn_wav = os.path.join('..', 'data', 'C8', 'FMP_C8_F10_Weber_Freischuetz-06_FreiDi-35-40.wav')\n",
"Fs = 22050\n",
"x, Fs = librosa.load(fn_wav, sr=Fs)\n",
"ipd.Audio(x, rate=Fs)\n",
"\n",
"# Computation of STFT\n",
"N = 2048\n",
"H = N//4\n",
"X = librosa.stft(x, n_fft=N, hop_length=H, win_length=N, pad_mode='constant')\n",
"Y = np.log(1 + np.abs(X))\n",
"Fs_feature = Fs/H\n",
"T_coef = np.arange(X.shape[1]) / Fs_feature\n",
"freq_res = Fs / N\n",
"F_coef = np.arange(X.shape[0]) * freq_res\n",
"#F_coef = libfmp.c3.F_coef(np.arange(Y.shape[0]), Fs, N)\n",
"\n",
"# Load F0 trajectory and and resample to fit to X\n",
"fn_traj = os.path.join('..', 'data', 'C8', 'FMP_C8_F10_Weber_Freischuetz-06_FreiDi-35-40_F0-user-Book.csv')\n",
"traj_df = libfmp.b.read_csv(fn_traj)\n",
"traj = traj_df.values\n",
"Fs_traj = 1/(traj[1,0]-traj[0,0])\n",
"traj_X_values = interp1d(traj[:,0], traj[:,1], kind='nearest', fill_value='extrapolate')(T_coef)\n",
"traj_X = np.hstack((T_coef[:,None], traj_X_values[:,None]))\n",
"\n",
"# Visualization\n",
"figsize=(7,4)\n",
"ylim = [0, 4000]\n",
"fig, ax = plot_STFT_F0_activity(Y, traj_X, figsize=figsize, ylim=ylim)\n",
"plt.show()\n",
"\n",
"# Sonification\n",
"soni_mono = libfmp.c8.sonify_trajectory_with_sinusoid(traj_X, len(x), Fs)\n",
"soni_stereo = np.vstack((x.reshape(1,-1), soni_mono.reshape(1,-1))) # left: x, right: sonification\n",
"ipd.Audio(soni_stereo, rate=Fs)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Melody Separation: Overall Idea\n",
"\n",
"In our simplified scenario, the objective of **melody extraction** is to estimate the sequence of F0-values that correspond to the main melody. A related task is referred to as **melody separation**, which aims at decomposing a music signal into a melody component that captures the main melodic voice and an accompaniment component that captures the remaining acoustic events. One application of melody separation is the automated generation of a **Karaoke version** for a given song, where the main melodic voice is to be removed from the song's original recording. More technically speaking, the objective of melody separation is to decompose a given signal $x$ into a melody component $x^\\mathrm{Mel}$ and an accompaniment component $x^\\mathrm{Acc}$ such that \n",
"\n",
"\\begin{equation}\n",
" x = x^\\mathrm{Mel} + x^\\mathrm{Acc}.\n",
"\\end{equation}\n",
"\n",
"One general approach is to first apply a melody extraction algorithm to derive the F0-trajectory of the main melody. Based on these **F0-values** and their **harmonics**, one can construct a **binary mask** for the melodic component, while the binary mask of the accompaniment is defined to be the complement. Using masking techniques as described in the [FMP notebook on HPS](../C8/C8S1_HPS.html) (see Section 8.1.1.2 of [Müller, FMP, Springer 2015]), one then derives the two masked STFTs $\\mathcal{X}^\\mathrm{Mel}$ and $\\mathcal{X}^\\mathrm{Acc}$. From this, one obtains the signals $x^\\mathrm{Mel}$ and $x^\\mathrm{Acc}$ by applying [signal reconstruction techniques](../C8/C8S1_SignalReconstruction.html) (see Section 8.1.2 of [Müller, FMP, Springer 2015]). We now implement this melody separation procedure and apply it to the Freischütz example. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Binary Mask\n",
"\n",
"There are many ways to define a **binary mask** for a given F0-trajectory. Taking harmonics into account, we provide in the following code cell two variants. \n",
"\n",
"* In the first variant, one can specify a tolerance parameters `tol_bin` (given in frequency bins), which is used to define a frequency neighborhood of **fixed size** around each frequency bin.\n",
"* In the second variant, one can specify a tolerance parameters `tol_cents` (given in cents), which is used to define a neighborhood of **frequency-dependent size** around each frequency bin. More precisely, the neighborhood size increases linearly with the center frequency (so that it remains constant when measured in cents).\n",
"\n",
"In the following figure, we show the binary masks for the melody (as well as their complements for the accompaniment) for both variants."
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"execution": {
"iopub.execute_input": "2024-02-15T08:43:50.182433Z",
"iopub.status.busy": "2024-02-15T08:43:50.182147Z",
"iopub.status.idle": "2024-02-15T08:43:50.768776Z",
"shell.execute_reply": "2024-02-15T08:43:50.768156Z"
}
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"
"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"#def display_signals_mel_acc(x_mel, x_acc, header_mel, header_acc):\n",
"# html_x_mel = libfmp.c8.generate_audio_tag_html_list([x_mel], Fs=Fs, width='350')\n",
"# html_x_acc = libfmp.c8.generate_audio_tag_html_list([x_acc], Fs=Fs, width='350')\n",
"#\n",
"# pd.set_option('display.max_colwidth', None) \n",
"# df = pd.DataFrame(OrderedDict([\n",
"# (header_mel, html_x_mel),\n",
"# (header_acc, html_x_acc)]))\n",
"# ipd.display(ipd.HTML(df.to_html(escape=False, header=True, index=False)))\n",
" \n",
"def compute_plot_display_mel_acc(Fs, X, N, H, mask_mel, figsize=figsize):\n",
" mask_acc = np.ones(mask_mel.shape) - mask_mel\n",
" X_mel = X * mask_mel\n",
" X_acc = X * mask_acc\n",
" x_mel = librosa.istft(X_mel, hop_length=H, win_length=N, window='hann', center=True, length=x.size)\n",
" x_acc = librosa.istft(X_acc, hop_length=H, win_length=N, window='hann', center=True, length=x.size)\n",
" \n",
" plt.figure(figsize=figsize)\n",
" ax = plt.subplot(121)\n",
" fig_im, ax_im, im = libfmp.b.plot_matrix(np.log(1+10 * np.abs(X_mel)), Fs=Fs_feature, ax=[ax], Fs_F=1/freq_res, figsize=figsize, \n",
" title='Masked STFT for melody');\n",
" plt.ylim(ylim);\n",
" ax = plt.subplot(122)\n",
" fig, ax, im = libfmp.b.plot_matrix(np.log(1+10 * np.abs(X_acc)), Fs=Fs_feature, ax=[ax], Fs_F=1/freq_res, figsize=figsize, \n",
" title='Masked STFT for accompaniment');\n",
" plt.ylim(ylim); \n",
" plt.show()\n",
" libfmp.b.audio_player_list([x_mel, x_acc], [Fs, Fs], width=350, \n",
" columns=['Reconstructed signal for melody', 'Reconstructed signal for accompaniment'])\n",
"\n",
"print('Melody separation (tol_bin = %d)'%tol_bin)\n",
"compute_plot_display_mel_acc(Fs, X, N, H, mask_mel_bin, figsize=figsize)\n",
"\n",
"print('Melody separation (tol_cent = %d)'%tol_cent)\n",
"compute_plot_display_mel_acc(Fs, X, N, H, mask_mel_cent, figsize=figsize)\n",
"\n",
"mask_mel_joint = np.logical_or(mask_mel_bin, mask_mel_cent)\n",
"print('Melody separation (joint mask using \"logical or\" operator)')\n",
"compute_plot_display_mel_acc(Fs, X, N, H, mask_mel_joint, figsize=figsize)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Example: Stop Messing With Me (Bornemark)\n",
"\n",
"Assuming that an F0-trajectory is given for the melody, the overall procedure for melody-accompaniment separation is summarized in the following code cell. As an example, we use excerpt of the song [\"Stop Messing With Me\" by Sven Bornemark](../C8/C8S2_FundFreqTracking.html), where the melody is performed by a male singer. As input to the separation procedure, we use the constraint-based F0-trajectory as computed at the end of the [FMP notebook on fundamental frequency tracking](../C8/C8S2_FundFreqTracking.html). Besides presenting the separated components signals, we also provide the ideal melody track and accompaniment track, which are also available for the song. \n",
"\n",
"\n",
"\n",
"\n",
""
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"execution": {
"iopub.execute_input": "2024-02-15T08:43:52.021640Z",
"iopub.status.busy": "2024-02-15T08:43:52.021424Z",
"iopub.status.idle": "2024-02-15T08:44:05.023162Z",
"shell.execute_reply": "2024-02-15T08:44:05.022388Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"