結果:
私の場合、前の会社が音楽認識アプリの会社で、アルゴリズム開発でFFTが使われていたことがきっかけでした。でも、MATLABのすごさが分かったのは、機械学習のオンライン講座で、Andrew Ngが、線型代数を使うと、数式と非常に近い構文のコードで問題が処理できることを学んだ時でした。
Three former MathWorks employees, Steve Wilcockson, David Bergstein, and Gareth Thomas, joined the ArrayCast pod cast to discuss their work on array based languages. At the end of the episode, Steve says,
> It's a little known fact about MATLAB. There's this thing, Gareth has talked about the community. One of the things MATLAB did very, very early was built the MATLAB community, the so-called MATLAB File Exchange, which came about in the early 2000s. And it was where people would share code sets, M files, et cetera. This was long before GitHub came around. This was well ahead of its time. And I think there are other places too, where MATLAB has delivered cultural benefits over and above the kind of core programming and mathematical capabilities too. So, you know, MATLAB Central, File Exchange, very much saw the future.
Listen here: The ArrayCast, Episode 79, May 10, 2024.
This topic is for discussing highlights to the current R2025a Pre-release.
So you've downloaded the R2025a pre-release, tried Dark mode and are wondering what else is new. A lot! A lot is new!
One thing I am particularly happy about is the fact that Apple Accelerate is now the default BLAS on Apple Silicon machines. Check it out by doing
>> version -blas
ans =
'Apple Accelerate BLAS (ILP64)'
If you compare this to R2024b that is using OpenBLAS you'll see some dramatic speed-ups in some areas. For example, I saw up to 3.7x speed-up for matrix-matrix multiplication on my M2 Mabook Pro and 2x faster LU factorisation.
Details regarding my experiments are in this blog post Life in the fast lane: Making MATLAB even faster on Apple Silicon with Apple Accelerate » The MATLAB Blog - MATLAB & Simulink . Back then you had to to some trickery to switch to Apple Accelerate, now its the default.
I just published a blog post called "The Story of TIMEIT." I've been thinking about writing something like this ever since Mike Croucher's tic/toc blog post last spring.
There were a lot of opinions about TIMEIT expressed in the comments of that blog post, including some of mine.
My blog post today gives a more full history of the function, its design goals, and how it works. I thought it might prompt more discussion, so I'm creating this thread as a place for it.
If you are an interested user of TIMEIT, feel free to weigh in here with your thoughts. Perhaps the thread will influence MathWorks regarding what to do with TIMEIT, or with related performance measurement capabilities.
Hi
If you have used the playground and are familiar with its capabilities, I will be very interested in your opinion about the tool.
Thank you in advance for your reply/opinion.
At
I am trying to run a scipt I from File Exchange, but I am getting error "Error using fsolve (line 186) FSOLVE requires at least two input arguments" Can please help fix the error?
% Prepare the options for the solver
options = prepareOptionsForSolver(options, 'fsolve');
end
if nargin == 0
error(message('optim:fsolve:NotEnoughInputs')) (line 186)
how can i open .lod extension and .iqd extension file in matlab?
Hi everyone
The R2025a pre-release is now available to licensed users. I highly encourage you to download, give it a try and give us some feedback.
The first thing I tried was switching to Dark mode. Here's the magic
>> s = settings;
>> s.matlab.appearance.MATLABTheme.PersonalValue = "Dark";
% 读取语音信号
try
[audio, fs] = audioread('floating.m4a');
% 输出读取到的音频数据的基本信息
fprintf('成功读取m4a文件，音频数据长度: %d，采样频率: %d\n', length(audio), fs);
catch ME
warning('无法读取m4a文件，尝试转换为wav格式');
try
[audio, fs] = audioread('floating.wav');
% 输出读取到的音频数据的基本信息
fprintf('成功读取wav文件，音频数据长度: %d，采样频率: %d\n', length(audio), fs);
catch ME2
error('无法读取任何格式的音频文件，错误信息: %s', ME2.message);
end
end
% 检查audio的数据类型
if ~isnumeric(audio)
audio = double(audio);
end
% 如果是多通道音频，转换为单通道
if size(audio, 2) > 1
audio = audio(:, 1);
end
% 对信号进行填充，使其长度为2的幂次方
len = length(audio);
next_pow2 = pow2(nextpow2(len));
if len < next_pow2
padding = zeros(next_pow2 - len, 1);
audio = [audio; padding];
end
% 检查audio是否为向量且非空
if ~isvector(audio) || isempty(audio)
error('audio必须是一个非空向量');
end
% 检查audio中的元素是否都是实数
if ~all(isreal(audio))
non_real_indices = find(~isreal(audio));
audio(non_real_indices) = real(audio(non_real_indices));
fprintf('检测到并处理了非实数元素\n');
end
% 再次确认audio的数据类型为double
if ~strcmp(class(audio), 'double')
audio = double(audio);
end
% 检查audio数据的统计信息
audio_mean = mean(audio);
audio_std = std(audio);
fprintf('音频数据的均值: %f，标准差: %f\n', audio_mean, audio_std);
% 调试输出audio的部分信息
fprintf('去噪前音频数据的前5个值: %f, %f, %f, %f, %f\n', audio(1), audio(2), audio(3), audio(4), audio(5));
% 1. 去噪
% 检查分解层数是否合理
max_decomposition_level = floor(log2(length(audio)));
if decomposition_level > max_decomposition_level
warning('设置的分解层数过高，将调整为最大可能层数 %d', max_decomposition_level);
decomposition_level = max_decomposition_level;
else
decomposition_level = 3;
end
% 检查小波函数、阈值规则等参数
valid_wavelets = {'sym4', 'db1', 'db2', 'haar'}; % 一些常见的有效小波函数
if ~ismember('sym4', valid_wavelets)
error('当前小波函数不被支持，请更换为有效的小波函数');
end
valid_threshold_rules = {'sqtwolog', 'rigrsure', 'heursure', 'minimaxi'}; % 一些常见的有效阈值规则
if ~ismember('sqtwolog', valid_threshold_rules)
error('当前阈值选择规则不被支持，请更换为有效的阈值规则');
end
% 尝试使用wdencmp函数进行小波去噪
% 这里设置KeepAPP为1，表示保留近似系数
[denoised_audio,~,~] = wdencmp('gbl', audio,'sym4', decomposition_level, 'sqtwolog', 'h', 1);
% 去除填充部分
denoised_audio = denoised_audio(1:len);
% 2. 归一化
normalized_audio = denoised_audio / max(abs(denoised_audio));
% 3. 加窗和分帧
window_size = 256; % 窗长
frame_shift = 128; % 帧移
window = hamming(window_size); % 汉明窗
% 分帧
num_frames = floor((length(normalized_audio) - window_size) / frame_shift) + 1;
frames = zeros(window_size, num_frames);
for i = 1:num_frames
start_index = (i - 1) * frame_shift + 1;
frames(:, i) = normalized_audio(start_index:start_index + window_size - 1).* window;
end
% 4. 频谱分析
% 使用spectrogram函数
noverlap = window_size - frame_shift;
[S, F, T] = spectrogram(normalized_audio, window, noverlap, num_fft, fs);
S = 10 * log10(abs(S));
% 5. 绘制频谱图
figure;
surf(T, F, S);
shading interp;
xlabel('Time (s)');
ylabel('Frequency (Hz)');
zlabel('Magnitude (dB)');
title('Spectrogram of Speech Signal');
>> Untitled3
成功读取m4a文件，音频数据长度: 479232，采样频率: 48000
音频数据的均值: 0.000000，标准差: 0.019026
去噪前音频数据的前5个值: 0.000000, 0.000000, 0.000000, 0.000000, 0.000000
***************************************
ARGUMENTS ERROR
---------------------------------------
wdencmp ---> real(s) => 0 , expected
***************************************
错误使用 wdencmp (line 91)
Invalid argument value.
出错 Untitled3 (line 83)
[denoised_audio,~,~] = wdencmp('gbl', audio,'sym4', decomposition_level, 'sqtwolog', 'h', 1);
The Most Accepted Badge and Top Downloads Badge are two prestigious annual honors that recognize outstanding contributions in MATLAB Answers and File Exchange.
Most Accepted badge is awarded to the top 10 contributors whose answers received the most acceptances. Top Downloads badge goes to the top 10 contributors with the highest number of downloads for their submissions. Please note that, starting in 2025, the criteria for Top Downloaded Badge in will be adjusted to only count downloads from files created or updated in 2025.
In 2024, the recipients for Most Accepted are: @Voss, @Walter Roberson, @Star Strider, @Torsten, @Matt J, @Stephen23, @Steven Lord, @Hassaan, @Sam Chak, and @Cris LaPierre.
The recipients for Top Downloaded are: @Steve Miller, @Rodney Tan, @Yair Altman, @Chad Greene, @William Thielicke, @Seyedali Mirjalili, @John D'Errico, @Giampiero Campa, @Toshiaki Takeuchi, and @Zhaoxu Liu / slandarer.
Congratulations and thank you again for your outstanding contributions in 2024!
Whether the computation of codes is still operating for matlab online after I close the website?
I have a skeleton and i want to find the start and end point of that skeleton. My skeleton is like an oval/donut where there is little gap between the start and end
Hello,
Now that the "Copilot+PC" (Windows ARM) laptops are rapidly increasing in market share (Microsoft Surface Laptop, Dell XPS 13, HP OmniBook X 14, and more), are there any plans to provide builds for Matlab on Windows arm64?
Since there are already Windows builds of Matlab, it shouldn't be too hard to compile for Windows arm64, as far as I know. But I am not famaliar with Matlab's codebase.
Please try to publish Windows arm64 builds soon so that Matlab can be much more usable on Windows on ARM as it will run natively instead of in emulation.
Thank you very much.
How does MATLAB app designer call external. m functions after packaging into exe? Please note that this external function is not packaged into EXE. Who can help me! I have tried commands such as' addpath ',' filread ',' eval ', but after testing, none of them worked.
for optimization is matlab better or python?
Hi!
I'm working on optical systems and I'm searching for algorithms and methods to solve the problem of autofocus.
I've already tried a bisection method and some optimization methods, but now I'd like to find a way to use neural networks in order to have a very fast and very precise way to move the lenses of my system to the right focus position.
Do you have any idea of the network I can use and how can I properly use it?
For the tests I'm using a setup made by an optical target, an objective and a sensor fixed on a linear positioner which is controlled by two functions I wrote in Matlab to move it back and forward.
Thank you!
I am trying to stimulate a given 3rd order chebyyshev filter on microwave transmission lines, with given impeadances of 3.34874, 0.71170, 3.34874
This is my code:
This is the stimulation result:
The LTSpice Stimulation result for same circuit:
I wonder why my Matlab stimulation have no Chebyshev ripple and different attenuation?
