-
Notifications
You must be signed in to change notification settings - Fork 2
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
28 changed files
with
1,552 additions
and
121 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Git LFS file not shown
Git LFS file not shown
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,79 @@ | ||
import librosa | ||
import numpy as np | ||
|
||
|
||
def readwav(filepath="61-70968-0000.wav", sr=16000): | ||
waveform, sr = librosa.load(filepath, sr=sr) | ||
|
||
print(f" - Length {len(waveform):6}") | ||
print(f" - Max {waveform.max():6.3f}") | ||
print(f" - Min {waveform.min():6.3f}") | ||
print(f" - Mean {waveform.mean():6.3f}") | ||
|
||
return waveform | ||
|
||
|
||
def pad_waveform(waveform): | ||
return np.concatenate([ | ||
np.zeros((512-400)//2), | ||
waveform, | ||
np.zeros((512-400)//2) | ||
]) | ||
|
||
|
||
def spectrogram(waveform, log_offset=1e-6): | ||
waveform = pad_waveform(waveform) | ||
|
||
S = librosa.stft( | ||
waveform, | ||
n_fft=512, | ||
hop_length=160, | ||
win_length=400, | ||
window="hann", | ||
center=False) | ||
S = np.log(np.abs(S) ** 2 + log_offset) | ||
|
||
return S.T.astype(np.float32) | ||
|
||
|
||
def melspectrogram(waveform, sr=16000, log_offset=1e-6): | ||
waveform = pad_waveform(waveform) | ||
|
||
M = librosa.feature.melspectrogram( | ||
y=waveform, | ||
sr=sr, | ||
n_fft=512, | ||
hop_length=160, | ||
win_length=400, | ||
window="hann", | ||
center=False, | ||
n_mels=64, | ||
htk=True, | ||
norm=None) | ||
M = np.log(M + log_offset) | ||
|
||
return M.T.astype(np.float32) | ||
|
||
|
||
def main(): | ||
waveform = readwav() | ||
|
||
print("Spectrogram") | ||
|
||
X = spectrogram(waveform) | ||
print(f" - Output {X.shape}") | ||
|
||
with open('spectrogram.bin', 'wb') as f: | ||
f.write(X.tobytes("C")) | ||
|
||
print("Mel-Spectrogram") | ||
|
||
X = melspectrogram(waveform) | ||
print(f" - Output {X.shape}") | ||
|
||
with open('melspectrogram.bin', 'wb') as f: | ||
f.write(X.tobytes("C")) | ||
|
||
|
||
if __name__ == "__main__": | ||
main() |
Binary file not shown.
Binary file not shown.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,76 @@ | ||
using Microsoft.VisualStudio.TestTools.UnitTesting; | ||
using System; | ||
using System.IO; | ||
using System.Runtime.InteropServices; | ||
|
||
namespace NeMoOnnxSharp.Tests | ||
{ | ||
[TestClass] | ||
public class FFTTest | ||
{ | ||
private static void CFFTRef(double[] xr, double[] xi, int N) | ||
{ | ||
double[] yr = new double[N]; | ||
double[] yi = new double[N]; | ||
for (int i = 0; i < N; i++) | ||
{ | ||
double vr = 0.0; | ||
double vi = 0.0; | ||
for (int k = 0; k < N; k++) | ||
{ | ||
vr += Math.Cos(-2 * Math.PI * k * i / N) * xr[k]; | ||
vi += Math.Sin(-2 * Math.PI * k * i / N) * xr[k]; | ||
} | ||
yr[i] = vr; | ||
yi[i] = vi; | ||
} | ||
for (int i = 0; i < N; i++) | ||
{ | ||
xr[i] = yr[i]; | ||
xi[i] = yi[i]; | ||
} | ||
} | ||
|
||
private static double MSE(double[] a, double[] b) | ||
{ | ||
if (a.Length != b.Length) throw new ArgumentException(); | ||
int len = Math.Min(a.Length, b.Length); | ||
double err = 0.0; | ||
for (int i = 0; i < len; i++) | ||
{ | ||
double diff = a[i] - b[i]; | ||
err += diff * diff; | ||
} | ||
return err / len; | ||
} | ||
|
||
[TestMethod] | ||
public void TestCFFT() | ||
{ | ||
var rng = new Random(); | ||
for (int N = 256; N <= 2048; N *= 2) | ||
{ | ||
var xr0 = new double[N]; | ||
var xi0 = new double[N]; | ||
var xr1 = new double[N]; | ||
var xi1 = new double[N]; | ||
for (int i = 0; i < 10; i++) | ||
{ | ||
for (int j = 0; j < N; j++) | ||
{ | ||
xr0[j] = rng.NextDouble(); | ||
xi0[j] = rng.NextDouble(); | ||
xr1[j] = xr0[j]; | ||
xi1[j] = rng.NextDouble(); | ||
} | ||
CFFTRef(xr0, xi0, N); | ||
FFT.CFFT(xr1, xi1, N); | ||
double error = MSE(xr0, xi1); | ||
Assert.IsTrue(error < 1e-20); | ||
error = MSE(xi0, xr1); | ||
Assert.IsTrue(error < 1e-20); | ||
} | ||
} | ||
} | ||
} | ||
} |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,40 @@ | ||
<Project Sdk="Microsoft.NET.Sdk"> | ||
|
||
<PropertyGroup> | ||
<TargetFramework>net7.0</TargetFramework> | ||
<Nullable>disable</Nullable> | ||
|
||
<IsPackable>false</IsPackable> | ||
</PropertyGroup> | ||
|
||
<ItemGroup> | ||
<PackageReference Include="Microsoft.NET.Test.Sdk" Version="17.5.0" /> | ||
<PackageReference Include="MSTest.TestAdapter" Version="3.0.2" /> | ||
<PackageReference Include="MSTest.TestFramework" Version="3.0.2" /> | ||
<PackageReference Include="coverlet.collector" Version="3.2.0"> | ||
<PrivateAssets>all</PrivateAssets> | ||
<IncludeAssets>runtime; build; native; contentfiles; analyzers; buildtransitive</IncludeAssets> | ||
</PackageReference> | ||
<PackageReference Include="System.Memory" Version="4.5.5" /> | ||
</ItemGroup> | ||
|
||
<ItemGroup> | ||
<ProjectReference Include="..\NeMoOnnxSharp\NeMoOnnxSharp.csproj" /> | ||
</ItemGroup> | ||
|
||
<ItemGroup> | ||
<None Update="Data\61-70968-0000-mod.wav"> | ||
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory> | ||
</None> | ||
<None Update="Data\61-70968-0000.wav"> | ||
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory> | ||
</None> | ||
<None Update="Data\melspectrogram.bin"> | ||
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory> | ||
</None> | ||
<None Update="Data\spectrogram.bin"> | ||
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory> | ||
</None> | ||
</ItemGroup> | ||
|
||
</Project> |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,160 @@ | ||
using Microsoft.VisualStudio.TestTools.UnitTesting; | ||
using System; | ||
using System.Diagnostics; | ||
using System.IO; | ||
using System.Reflection; | ||
using System.Runtime.InteropServices; | ||
|
||
namespace NeMoOnnxSharp.Tests | ||
{ | ||
[TestClass] | ||
public class NeMoOnnxSharpTest | ||
{ | ||
private const int SampleRate = 16000; | ||
private const string SampleWAVSpeechFile = "61-70968-0000.wav"; | ||
|
||
private static float[] ReadData(string file) | ||
{ | ||
string appDirPath = AppDomain.CurrentDomain.BaseDirectory; | ||
string path = Path.Combine(appDirPath, "Data", file); | ||
var bytes = File.ReadAllBytes(path); | ||
return MemoryMarshal.Cast<byte, float>(bytes).ToArray(); | ||
} | ||
|
||
private static double MSE(float[] a, float[] b) | ||
{ | ||
if (a.Length != b.Length) throw new ArgumentException(); | ||
int len = Math.Min(a.Length, b.Length); | ||
double err = 0.0; | ||
for (int i = 0; i < len; i++) | ||
{ | ||
double diff = a[i] - b[i]; | ||
err += diff * diff; | ||
} | ||
return err / len; | ||
} | ||
|
||
private static double MSE(double[] a, double[] b) | ||
{ | ||
if (a.Length != b.Length) throw new ArgumentException(); | ||
int len = Math.Min(a.Length, b.Length); | ||
double err = 0.0; | ||
for (int i = 0; i < len; i++) | ||
{ | ||
double diff = a[i] - b[i]; | ||
err += diff * diff; | ||
} | ||
return err / len; | ||
} | ||
|
||
short[] waveform; | ||
AudioProcessor processor; | ||
|
||
public NeMoOnnxSharpTest() | ||
{ | ||
string appDirPath = AppDomain.CurrentDomain.BaseDirectory; | ||
string waveFile = Path.Combine(appDirPath, "Data", SampleWAVSpeechFile); | ||
waveform = WaveFile.ReadWAV(waveFile, SampleRate); | ||
processor = new AudioProcessor( | ||
sampleRate: SampleRate, | ||
window: "hann", | ||
windowLength: 400, | ||
hopLength: 160, | ||
fftLength: 512, | ||
//preNormalize: 0.8, | ||
preemph: 0.0, | ||
center: false, | ||
nMelBands: 64, | ||
melMinHz: 0.0, | ||
melMaxHz: 0.0, | ||
htk: true, | ||
melNormalize: null, | ||
logOffset: 1e-6, | ||
postNormalize: false); | ||
} | ||
|
||
[TestMethod] | ||
public void TestSpectrogram() | ||
{ | ||
var x = processor.Spectrogram(waveform); | ||
AssertMSE("spectrogram.bin", x); | ||
} | ||
|
||
[TestMethod] | ||
public void TestMelSpectrogram() | ||
{ | ||
var x = processor.MelSpectrogram(waveform); | ||
AssertMSE("melspectrogram.bin", x); | ||
} | ||
|
||
[TestMethod] | ||
public void TestReadFrame() | ||
{ | ||
int windowLength = 5; | ||
int fftLength = 9; | ||
var processor = new AudioProcessor( | ||
windowLength: windowLength, | ||
fftLength: fftLength, | ||
preemph: 0.0); | ||
|
||
MethodInfo methodInfo1 = typeof(AudioProcessor).GetMethod( | ||
"ReadFrameCenter", BindingFlags.NonPublic | BindingFlags.Instance); | ||
MethodInfo methodInfo2 = typeof(AudioProcessor).GetMethod( | ||
"ReadFrameCenterPreemphasis", BindingFlags.NonPublic | BindingFlags.Instance); | ||
|
||
var rng = new Random(); | ||
short[] waveform = new short[1200]; | ||
double[] frame1 = new double[fftLength]; | ||
double[] frame2 = new double[fftLength]; | ||
for (int i = 0; i < waveform.Length; i++) waveform[i] = (short)rng.Next(short.MinValue, short.MaxValue); | ||
#if true | ||
for (int i = 0; i < 100; i++) | ||
{ | ||
int offset = rng.Next(waveform.Length); | ||
double scale = rng.NextDouble(); | ||
object[] parameters1 = { waveform, offset, scale, frame1 }; | ||
methodInfo1.Invoke(processor, parameters1); | ||
object[] parameters2 = { waveform, offset, scale, frame2 }; | ||
methodInfo2.Invoke(processor, parameters2); | ||
double error = MSE(frame1, frame2); | ||
Assert.IsTrue(error == 0); | ||
} | ||
#else | ||
for (int j = 0; j < 5; j++) | ||
{ | ||
var stopWatch = new Stopwatch(); | ||
stopWatch.Start(); | ||
for (int i = 0; i < 1000000; i++) | ||
{ | ||
int offset = rng.Next(waveform.Length); | ||
double scale = rng.NextDouble(); | ||
object[] parameters1 = { waveform, offset, scale, frame1 }; | ||
methodInfo1.Invoke(processor, parameters1); | ||
} | ||
stopWatch.Stop(); | ||
Console.WriteLine(stopWatch.Elapsed); | ||
|
||
stopWatch = new Stopwatch(); | ||
stopWatch.Start(); | ||
for (int i = 0; i < 1000000; i++) | ||
{ | ||
int offset = rng.Next(waveform.Length); | ||
double scale = rng.NextDouble(); | ||
object[] parameters2 = { waveform, offset, scale, frame1 }; | ||
methodInfo2.Invoke(processor, parameters2); | ||
} | ||
stopWatch.Stop(); | ||
Console.WriteLine(stopWatch.Elapsed); | ||
} | ||
#endif | ||
} | ||
|
||
private void AssertMSE(string path, float[] x, double threshold = 1e-3) | ||
{ | ||
var truth = ReadData(path); | ||
double mse = MSE(truth, x); | ||
Console.WriteLine("MSE: {0}", mse); | ||
Assert.IsTrue(mse < threshold); | ||
} | ||
} | ||
} |
Oops, something went wrong.