OLD | NEW |
| (Empty) |
1 /* | |
2 * Copyright (c) 2014 The WebRTC project authors. All Rights Reserved. | |
3 * | |
4 * Use of this source code is governed by a BSD-style license | |
5 * that can be found in the LICENSE file in the root of the source | |
6 * tree. An additional intellectual property rights grant can be found | |
7 * in the file PATENTS. All contributing project authors may | |
8 * be found in the AUTHORS file in the root of the source tree. | |
9 */ | |
10 | |
11 #include "webrtc/modules/audio_coding/neteq/audio_classifier.h" | |
12 | |
13 #include <math.h> | |
14 #include <stdio.h> | |
15 #include <stdlib.h> | |
16 #include <string.h> | |
17 | |
18 #include <iostream> | |
19 #include <memory> | |
20 #include <string> | |
21 | |
22 int main(int argc, char* argv[]) { | |
23 if (argc != 5) { | |
24 std::cout << "Usage: " << argv[0] << | |
25 " channels output_type <input file name> <output file name> " | |
26 << std::endl << std::endl; | |
27 std::cout << "Where channels can be 1 (mono) or 2 (interleaved stereo),"; | |
28 std::cout << " outputs can be 1 (classification (boolean)) or 2"; | |
29 std::cout << " (classification and music probability (float))," | |
30 << std::endl; | |
31 std::cout << "and the sampling frequency is assumed to be 48 kHz." | |
32 << std::endl; | |
33 return -1; | |
34 } | |
35 | |
36 const int kFrameSizeSamples = 960; | |
37 int channels = atoi(argv[1]); | |
38 if (channels < 1 || channels > 2) { | |
39 std::cout << "Disallowed number of channels " << channels << std::endl; | |
40 return -1; | |
41 } | |
42 | |
43 int outputs = atoi(argv[2]); | |
44 if (outputs < 1 || outputs > 2) { | |
45 std::cout << "Disallowed number of outputs " << outputs << std::endl; | |
46 return -1; | |
47 } | |
48 | |
49 const int data_size = channels * kFrameSizeSamples; | |
50 std::unique_ptr<int16_t[]> in(new int16_t[data_size]); | |
51 | |
52 std::string input_filename = argv[3]; | |
53 std::string output_filename = argv[4]; | |
54 | |
55 std::cout << "Input file: " << input_filename << std::endl; | |
56 std::cout << "Output file: " << output_filename << std::endl; | |
57 | |
58 FILE* in_file = fopen(input_filename.c_str(), "rb"); | |
59 if (!in_file) { | |
60 std::cout << "Cannot open input file " << input_filename << std::endl; | |
61 return -1; | |
62 } | |
63 | |
64 FILE* out_file = fopen(output_filename.c_str(), "wb"); | |
65 if (!out_file) { | |
66 std::cout << "Cannot open output file " << output_filename << std::endl; | |
67 return -1; | |
68 } | |
69 | |
70 webrtc::AudioClassifier classifier; | |
71 int frame_counter = 0; | |
72 int music_counter = 0; | |
73 while (fread(in.get(), sizeof(*in.get()), | |
74 data_size, in_file) == (size_t) data_size) { | |
75 bool is_music = classifier.Analysis(in.get(), data_size, channels); | |
76 if (!fwrite(&is_music, sizeof(is_music), 1, out_file)) { | |
77 std::cout << "Error writing." << std::endl; | |
78 return -1; | |
79 } | |
80 if (is_music) { | |
81 music_counter++; | |
82 } | |
83 std::cout << "frame " << frame_counter << " decision " << is_music; | |
84 if (outputs == 2) { | |
85 float music_prob = classifier.music_probability(); | |
86 if (!fwrite(&music_prob, sizeof(music_prob), 1, out_file)) { | |
87 std::cout << "Error writing." << std::endl; | |
88 return -1; | |
89 } | |
90 std::cout << " music prob " << music_prob; | |
91 } | |
92 std::cout << std::endl; | |
93 frame_counter++; | |
94 } | |
95 std::cout << frame_counter << " frames processed." << std::endl; | |
96 if (frame_counter > 0) { | |
97 float music_percentage = music_counter / static_cast<float>(frame_counter); | |
98 std::cout << music_percentage << " percent music." << std::endl; | |
99 } | |
100 | |
101 fclose(in_file); | |
102 fclose(out_file); | |
103 return 0; | |
104 } | |
OLD | NEW |