与《分享用于学习C++图像处理的代码示例》为姊妹篇。
为了便于学习C++音频处理并研究音频算法,
俺写了一个适合初学者学习的小小框架。
麻雀虽小五脏俱全,仅仅考虑单通道处理。
采用Decoder:dr_wav
https://github.com/mackron/dr_libs/blob/master/dr_wav.h
采用Encoder:原本计划采用dr_wav的Encode,但是dr_wav保存的文件头忘记修正音频数据的大小,
采用博主自己实现的代码,仅供学习之用。
dr_wav用于解析wav文件格式.
关于wav格式的解析移步至:
http://soundfile.sapp.org/doc/WaveFormat/
个人习惯,采用int16的处理方式,也可以通过简单的修改,改为float类型。
wav音频样本可以从*上(https://en.wikipedia.org/wiki/WAV)下载。
注:少数wav格式不支持
Format | Bitrate (kbit/s) | 1 minute (KiB) | Sample |
---|---|---|---|
11,025 Hz 16 bit PCM | 176.4 | 1292 | 11k16bitpcm.wav |
8,000 Hz 16 bit PCM | 128 | 938 | 8k16bitpcm.wav |
11,025 Hz 8 bit PCM | 88.2 | 646 | 11k8bitpcm.wav |
11,025 Hz µ-Law | 88.2 | 646 | 11kulaw.wav |
8,000 Hz 8 bit PCM | 64 | 469 | 8k8bitpcm.wav |
8,000 Hz µ-Law | 64 | 469 | 8kulaw.wav |
11,025 Hz 4 bit ADPCM | 44.1 | 323 | 11kadpcm.wav |
8,000 Hz 4 bit ADPCM | 32 | 234 | 8kadpcm.wav |
11,025 Hz GSM 06.10 | 18 | 132 | 11kgsm.wav |
8,000 Hz MP3 16 kbit/s | 16 | 117 | 8kmp316.wav |
8,000 Hz GSM 06.10 | 13 | 103 | 8kgsm.wav |
8,000 Hz Lernout & Hauspie SBC 12 kbit/s | 12 | 88 | 8ksbc12.wav |
8,000 Hz DSP Group Truespeech | 9 | 66 | 8ktruespeech.wav |
8,000 Hz MP3 8 kbit/s | 8 | 60 | 8kmp38.wav |
8,000 Hz Lernout & Hauspie CELP | 4.8 | 35 | 8kcelp.wav |
附带处理耗时计算,示例演示了一个简单的将音频前面一半静音处理,并简单注释了一下部分逻辑。
完整代码:
#include <stdio.h> #include <stdlib.h> #include <stdint.h> #include <time.h> #include <iostream> //采用https://github.com/mackron/dr_libs/blob/master/dr_wav.h 解码 #define DR_WAV_IMPLEMENTATION #include "dr_wav.h" auto const epoch = clock(); static double now() { return (clock() - epoch); }; template <typename FN> static double bench(const FN &fn) { auto took = -now(); ; } //写wav文件 void wavWrite_int16(char* filename, int16_t* buffer, int sampleRate, uint32_t totalSampleCount) { FILE* fp = fopen(filename, "wb"); if (fp == NULL) { printf("文件打开失败.\n"); return; } //修正写入的buffer长度 totalSampleCount *= sizeof(int16_t); ; ; ; ] = { 'R', 'I', 'F', 'F' }; uint32_t long_number = + totalSampleCount; fwrite(text, , , fp); fwrite(&long_number, , , fp); text[] = 'W'; text[] = 'A'; text[] = 'V'; text[] = 'E'; fwrite(text, , , fp); text[] = 'f'; text[] = 'm'; text[] = 't'; text[] = ' '; fwrite(text, , , fp); long_number = ; fwrite(&long_number, , , fp); int16_t short_number = FORMAT_PCM;//默认音频格式 fwrite(&short_number, , , fp); short_number = ; // 音频通道数 fwrite(&short_number, , , fp); long_number = sampleRate; // 采样率 fwrite(&long_number, , , fp); long_number = sampleRate * nbyte; // 比特率 fwrite(&long_number, , , fp); short_number = nbyte; // 块对齐 fwrite(&short_number, , , fp); short_number = nbit; // 采样精度 fwrite(&short_number, , , fp); ] = { 'd', 'a', 't', 'a' }; fwrite(data, , , fp); long_number = totalSampleCount; fwrite(&long_number, , , fp); fwrite(buffer, totalSampleCount, , fp); fclose(fp); } //读取wav文件 int16_t* wavRead_int16(char* filename, uint32_t* sampleRate, uint64_t *totalSampleCount) { unsigned int channels; int16_t* buffer = drwav_open_and_read_file_s16(filename, &channels, sampleRate, totalSampleCount); if (buffer == NULL) { printf("读取wav文件失败."); } //仅仅处理单通道音频 ) { drwav_free(buffer); buffer = NULL; *sampleRate = ; *totalSampleCount = ; } return buffer; } //分割路径函数 void splitpath(const char* path, char* drv, char* dir, char* name, char* ext) { const char* end; const char* p; const char* s; ] && path[] == ':') { if (drv) { *drv++ = *path++; *drv++ = *path++; *drv = '\0'; } } else if (drv) *drv = '\0'; for (end = path; *end && *end != ':';) end++; for (p = end; p > path && *--p != '\\' && *p != '/';) if (*p == '.') { end = p; break; } if (ext) for (s = end; (*ext = *s++);) ext++; for (p = end; p > path;) if (*--p == '\\' || *p == '/') { p++; break; } if (name) { for (s = p; s < end;) *name++ = *s++; *name = '\0'; } if (dir) { for (s = path; s < p;) *dir++ = *s++; *dir = '\0'; } } int main(int argc, char* argv[]) { std::cout << "Audio Processing " << std::endl; std::cout << "博客:http://tntmonks.cnblogs.com/" << std::endl; std::cout << "支持解析单通道wav格式." << std::endl; ) ; ]; //音频采样率 uint32_t sampleRate = ; //总音频采样数 uint64_t totalSampleCount = ; int16_t* wavBuffer = NULL; double nLoadTime = bench([&] { wavBuffer = wavRead_int16(in_file, &sampleRate, &totalSampleCount); }); std::cout << ) << " 毫秒" << std::endl; //如果加载成功 if (wavBuffer != NULL) { //将前面一般进行静音处理,直接置零即可 ; i < totalSampleCount / ; i++) { wavBuffer[i] = ; } } //保存结果 double nSaveTime = bench([&] { ]; ]; ]; ]; ]; splitpath(in_file, drive, dir, fname, ext); sprintf(out_file, "%s%s%s_out%s", drive, dir, fname, ext); wavWrite_int16(out_file, wavBuffer, sampleRate, totalSampleCount); }); std::cout << ) << " 毫秒" << std::endl; getchar(); std::cout << "按任意键退出程序 \n" << std::endl; ; }
示例具体流程为:
加载wav(拖放wav文件到可执行文件上)->简单静音处理->保存wav
并对 加载,保存 这2个环节都进行了耗时计算并输出。
若有其他相关问题或者需求也可以邮件联系俺探讨。
邮箱地址是:
gaozhihan@vip.qq.com
若此博文能帮到您,欢迎扫码小额赞助。
微信:
支付宝: