VampPluginSDK 2.10
PercussionOnsetDetector.cpp
Go to the documentation of this file.
1/* -*- c-basic-offset: 4 indent-tabs-mode: nil -*- vi:set ts=8 sts=4 sw=4: */
2
3/*
4 Vamp
5
6 An API for audio analysis and feature extraction plugins.
7
8 Centre for Digital Music, Queen Mary, University of London.
9 Copyright 2006 Chris Cannam.
10
11 Permission is hereby granted, free of charge, to any person
12 obtaining a copy of this software and associated documentation
13 files (the "Software"), to deal in the Software without
14 restriction, including without limitation the rights to use, copy,
15 modify, merge, publish, distribute, sublicense, and/or sell copies
16 of the Software, and to permit persons to whom the Software is
17 furnished to do so, subject to the following conditions:
18
19 The above copyright notice and this permission notice shall be
20 included in all copies or substantial portions of the Software.
21
22 THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
23 EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
24 MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
25 NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR
26 ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
27 CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
28 WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
29
30 Except as contained in this notice, the names of the Centre for
31 Digital Music; Queen Mary, University of London; and Chris Cannam
32 shall not be used in advertising or otherwise to promote the sale,
33 use or other dealings in this Software without prior written
34 authorization.
35*/
36
38
39using std::string;
40using std::vector;
41using std::cerr;
42using std::endl;
43
44#include <cmath>
45
46
48 Plugin(inputSampleRate),
49 m_stepSize(0),
50 m_blockSize(0),
51 m_threshold(3),
52 m_sensitivity(40),
53 m_priorMagnitudes(0),
54 m_dfMinus1(0),
55 m_dfMinus2(0)
56{
57}
58
60{
61 delete[] m_priorMagnitudes;
62}
63
64string
66{
67 return "percussiononsets";
68}
69
70string
72{
73 return "Simple Percussion Onset Detector";
74}
75
76string
78{
79 return "Detect percussive note onsets by identifying broadband energy rises";
80}
81
82string
84{
85 return "Vamp SDK Example Plugins";
86}
87
88int
90{
91 return 2;
92}
93
94string
96{
97 return "Code copyright 2006 Queen Mary, University of London, after Dan Barry et al 2005. Freely redistributable (BSD license)";
98}
99
100size_t
102{
103 return 0;
104}
105
106size_t
108{
109 return 1024;
110}
111
112bool
113PercussionOnsetDetector::initialise(size_t channels, size_t stepSize, size_t blockSize)
114{
115 if (channels < getMinChannelCount() ||
116 channels > getMaxChannelCount()) return false;
117
118 m_stepSize = stepSize;
119 m_blockSize = blockSize;
120
121 m_priorMagnitudes = new float[m_blockSize/2];
122
123 for (size_t i = 0; i < m_blockSize/2; ++i) {
124 m_priorMagnitudes[i] = 0.f;
125 }
126
127 m_dfMinus1 = 0.f;
128 m_dfMinus2 = 0.f;
129
130 return true;
131}
132
133void
135{
136 for (size_t i = 0; i < m_blockSize/2; ++i) {
137 m_priorMagnitudes[i] = 0.f;
138 }
139
140 m_dfMinus1 = 0.f;
141 m_dfMinus2 = 0.f;
142}
143
146{
147 ParameterList list;
148
150 d.identifier = "threshold";
151 d.name = "Energy rise threshold";
152 d.description = "Energy rise within a frequency bin necessary to count toward broadband total";
153 d.unit = "dB";
154 d.minValue = 0;
155 d.maxValue = 20;
156 d.defaultValue = 3;
157 d.isQuantized = false;
158 list.push_back(d);
159
160 d.identifier = "sensitivity";
161 d.name = "Sensitivity";
162 d.description = "Sensitivity of peak detector applied to broadband detection function";
163 d.unit = "%";
164 d.minValue = 0;
165 d.maxValue = 100;
166 d.defaultValue = 40;
167 d.isQuantized = false;
168 list.push_back(d);
169
170 return list;
171}
172
173float
175{
176 if (id == "threshold") return m_threshold;
177 if (id == "sensitivity") return m_sensitivity;
178 return 0.f;
179}
180
181void
182PercussionOnsetDetector::setParameter(std::string id, float value)
183{
184 if (id == "threshold") {
185 if (value < 0) value = 0;
186 if (value > 20) value = 20;
187 m_threshold = value;
188 } else if (id == "sensitivity") {
189 if (value < 0) value = 0;
190 if (value > 100) value = 100;
191 m_sensitivity = value;
192 }
193}
194
197{
198 OutputList list;
199
201 d.identifier = "onsets";
202 d.name = "Onsets";
203 d.description = "Percussive note onset locations";
204 d.unit = "";
205 d.hasFixedBinCount = true;
206 d.binCount = 0;
207 d.hasKnownExtents = false;
208 d.isQuantized = false;
209 d.sampleType = OutputDescriptor::VariableSampleRate;
211 list.push_back(d);
212
213 d.identifier = "detectionfunction";
214 d.name = "Detection Function";
215 d.description = "Broadband energy rise detection function";
216 d.binCount = 1;
217 d.isQuantized = true;
218 d.quantizeStep = 1.0;
219 d.sampleType = OutputDescriptor::OneSamplePerStep;
220 list.push_back(d);
221
222 return list;
223}
224
226PercussionOnsetDetector::process(const float *const *inputBuffers,
228{
229 if (m_stepSize == 0) {
230 cerr << "ERROR: PercussionOnsetDetector::process: "
231 << "PercussionOnsetDetector has not been initialised"
232 << endl;
233 return FeatureSet();
234 }
235
236 int count = 0;
237
238 for (size_t i = 1; i < m_blockSize/2; ++i) {
239
240 float real = inputBuffers[0][i*2];
241 float imag = inputBuffers[0][i*2 + 1];
242
243 float sqrmag = real * real + imag * imag;
244
245 if (m_priorMagnitudes[i] > 0.f) {
246 float diff = 10.f * log10f(sqrmag / m_priorMagnitudes[i]);
247
248// std::cout << "i=" << i << ", sqrmag=" << sqrmag << ", prior=" << m_priorMagnitudes[i] << ", diff=" << diff << ", threshold=" << m_threshold << " " << (diff >= m_threshold ? "[*]" : "") << std::endl;
249
250 if (diff >= m_threshold) ++count;
251 }
252
253 m_priorMagnitudes[i] = sqrmag;
254 }
255
256 FeatureSet returnFeatures;
257
258 Feature detectionFunction;
259 detectionFunction.hasTimestamp = false;
260 detectionFunction.values.push_back(count);
261 returnFeatures[1].push_back(detectionFunction);
262
263 if (m_dfMinus2 < m_dfMinus1 &&
264 m_dfMinus1 >= count &&
265 m_dfMinus1 > ((100 - m_sensitivity) * m_blockSize) / 200) {
266
267//std::cout << "result at " << ts << "! (count == " << count << ", prev == " << m_dfMinus1 << ")" << std::endl;
268
269 Feature onset;
270 onset.hasTimestamp = true;
272 (m_stepSize, int(m_inputSampleRate + 0.5));
273 returnFeatures[0].push_back(onset);
274 }
275
277 m_dfMinus1 = count;
278
279 return returnFeatures;
280}
281
284{
285 return FeatureSet();
286}
287
size_t getPreferredStepSize() const
Get the preferred step size (window increment – the distance in sample frames between the start frame...
std::string getMaker() const
Get the name of the author or vendor of the plugin in human-readable form.
bool initialise(size_t channels, size_t stepSize, size_t blockSize)
Initialise a plugin to prepare it for use with the given number of input channels,...
std::string getDescription() const
Get a human-readable description for the plugin, typically a line of text that may optionally be disp...
OutputList getOutputDescriptors() const
Get the outputs of this plugin.
void reset()
Reset the plugin after use, to prepare it for another clean run.
std::string getName() const
Get a human-readable name or title of the plugin.
PercussionOnsetDetector(float inputSampleRate)
void setParameter(std::string id, float value)
Set a named parameter.
size_t getPreferredBlockSize() const
Get the preferred block size (window size – the number of sample frames passed in each block to the p...
std::string getIdentifier() const
Get the computer-usable name of the plugin.
float getParameter(std::string id) const
Get the value of a named parameter.
std::string getCopyright() const
Get the copyright statement or licensing summary for the plugin.
FeatureSet getRemainingFeatures()
After all blocks have been processed, calculate and return any remaining features derived from the co...
int getPluginVersion() const
Get the version number of the plugin.
FeatureSet process(const float *const *inputBuffers, Vamp::RealTime timestamp)
Process a single block of input data.
ParameterList getParameterDescriptors() const
Get the controllable parameters of this plugin.
std::vector< ParameterDescriptor > ParameterList
virtual size_t getMaxChannelCount() const
Get the maximum supported number of input channels.
std::vector< OutputDescriptor > OutputList
std::map< int, FeatureList > FeatureSet
float m_inputSampleRate
virtual size_t getMinChannelCount() const
Get the minimum supported number of input channels.
std::string name
The human-readable name of the parameter.
std::string identifier
The name of the parameter, in computer-usable form.
std::string unit
The unit of the parameter, in human-readable form.
std::string description
A human-readable short text describing the parameter.
bool isQuantized
True if the parameter values are quantized to a particular resolution.
float minValue
The minimum value of the parameter.
float maxValue
The maximum value of the parameter.
float defaultValue
The default value of the parameter.
std::vector< float > values
Results for a single sample of this feature.
bool hasTimestamp
True if an output feature has its own timestamp.
RealTime timestamp
Timestamp of the output feature.
SampleType sampleType
Positioning in time of the output results.
bool isQuantized
True if the output values are quantized to a particular resolution.
float quantizeStep
Quantization resolution of the output values (e.g.
std::string unit
The unit of the output, in human-readable form.
std::string name
The human-readable name of the output.
float sampleRate
Sample rate of the output results, as samples per second.
std::string identifier
The name of the output, in computer-usable form.
bool hasKnownExtents
True if the results in each output bin fall within a fixed numeric range (minimum and maximum values)...
bool hasFixedBinCount
True if the output has the same number of values per sample for every output sample.
std::string description
A human-readable short text describing the output.
size_t binCount
The number of values per result of the output.
RealTime represents time values to nanosecond precision with accurate arithmetic and frame-rate conve...
static RealTime frame2RealTime(long frame, unsigned int sampleRate)
Convert a sample frame at the given sample rate into a RealTime.