Go语学习笔记 - 调用ffmpeg-api实现音频重采样_go ffmpeg api

目录

前言

环境部署

代码

总结

前言

最近对golang处理音视频很感兴趣，对golang音视频常用库goav进行了一番研究。自己写了一个wav转采样率的功能。给大家分享一下，中间遇到了不少坑，解决的过程中还是蛮有意思的。

环境部署

代码运行在Ubuntu环境上，需要使用到goav，goav是对ffmpeg源码的golang封装。

goav地址：https://github.com/giorgisio/goav

goav安装如下

sudo apt-get -y install autoconf automake build-essential libass-dev libfreetype6-dev libsdl1.2-dev libtheora-dev libtool libva-dev libvdpau-dev libvorbis-dev libxcb1-dev libxcb-shm0-dev libxcb-xfixes0-dev pkg-config texi2html zlib1g-dev
 
sudo apt install -y libavdevice-dev libavfilter-dev libswscale-dev libavcodec-dev libavformat-dev libswresample-dev libavutil-dev
 
sudo apt-get install yasm
 
export FFMPEG_ROOT=$HOME/ffmpeg
export CGO_LDFLAGS="-L$FFMPEG_ROOT/lib/ -lavcodec -lavformat -lavutil -lswscale -lswresample -lavdevice -lavfilter"
export CGO_CFLAGS="-I$FFMPEG_ROOT/include"
export LD_LIBRARY_PATH=$HOME/ffmpeg/lib
``` 
 
``` 
go get github.com/xueqing/goav

代码

先看代码

package main
 
//#include<stdlib.h>
import "C"
import (
	"flag"
	"fmt"
	"github.com/google/logger"
	"github.com/xueqing/ffmpeg-demo/logutil"
	"github.com/xueqing/goav/libswresample"
	"github.com/youpy/go-wav"
	"io"
	"os"
	"reflect"
	"unsafe"
)
 
func main() {
	var (
		inputUrl      string = "./data/1.wav"
		inNumChannels int64  = 1
		inSampleRate  int    = 16000
		//inBitsPerSample  uint16                    = 16
		outNumChannels   int64                     = 1
		outSampleRate    int                       = 48000
		outBitsPerSample uint16                    = 16
		swr              *libswresample.SwrContext = libswresample.SwrAlloc()
	)
	flag.Parse()
	logutil.Init(true, false, "resample.log")
	defer logutil.Close()
	swr.SwrAllocSetOpts(outNumChannels,
		libswresample.AvSampleFormat(1),
		outSampleRate,
		inNumChannels,
		libswresample.AvSampleFormat(1),
		inSampleRate,
		0,
		0)
	swr.SwrInit()
	defer swr.SwrClose()
 
	_inputFile, err := os.Open(inputUrl)
	if err != nil {
		logger.Errorf("open input file error(%v)", err)
		return
	}
	defer _inputFile.Close()
	_reader := wav.NewReader(_inputFile)
	format, err := _reader.Format()
	if err != nil {
		logger.Errorf("input file format error(%v)", err)
		return
	}
	fmt.Printf("input file format info -> AudioFormat:%v,NumChannels:%v,SampleRate:%v,ByteRate:%v,BlockAlign:%v,BitsPerSample:%v", int(format.AudioFormat), format.NumChannels, format.SampleRate, format.ByteRate, format.BlockAlign, format.BitsPerSample)
 
	_tempFile, err := os.CreateTemp("", "*.wav")
	if err != nil {
		logger.Errorf("create temp file error(%v)", err)
		return
	}
	logger.Infof("Create tempFile %v", _tempFile.Name())
	defer func() {
		_tempFile.Close()
	}()
	_samples := []wav.Sample{}
	n := 4096
	for {
		spls, err := _reader.ReadSamples(uint32(n))
		if err == io.EOF {
			break
		}
		_samples = append(_samples, spls...)
	}
	_result := ResampleByFFmpegApi2(swr, _samples)
	_writer := wav.NewWriter(_tempFile, uint32(len(_result)), uint16(outNumChannels), uint32(outSampleRate), outBitsPerSample)
 
	err4 := _writer.WriteSamples(_result)
	if err4 != nil {
		logger.Errorf("write file error(%v)", err4)
		err = err4
		return
	}
}
 
func ResampleByFFmpegApi2(swr *libswresample.SwrContext, samples []wav.Sample) []wav.Sample {
	var (
		_inArr  **uint8
		_outArr **uint8
		_inptr  []uint16
		_outptr []uint16
	)
	_inArr = (**uint8)(C.malloc(C.sizeof_int))
	defer C.free(unsafe.Pointer(_inArr))
	_inptr = make([]uint16, len(samples))
	_outArr = (**uint8)(C.malloc(C.sizeof_int))
	defer C.free(unsafe.Pointer(_outArr))
	_outptr = make([]uint16, len(samples)*3)
	//fmt.Println(unsafe.Sizeof(uint16(0)))
	for i, v := range samples {
		_inptr[i] = uint16(v.Values[0])
	}
	*_inArr = (*uint8)(unsafe.Pointer((*reflect.SliceHeader)(unsafe.Pointer(&_inptr)).Data))
	*_outArr = (*uint8)(unsafe.Pointer((*reflect.SliceHeader)(unsafe.Pointer(&_outptr)).Data))
	ret := swr.SwrConvert(_outArr, len(samples)*3, _inArr, len(samples))
	if ret > 0 {
		fmt.Println(ret)
	}
	_result := make([]wav.Sample, ret)
 
	for i := 0; i < ret; i++ {
		_result[i] = wav.Sample{[2]int{int(_outptr[i]), 0}}
	}
	return _result
}

代码说明：

1、代码不是个工具方法，如果看懂逻辑的话，可以自行修改成工具方法。

2、里面会用到ffmpeg里面swresample库，对音频数据进行冲采样。

3、可以细看一下，如果你想作实时处理也是可以改的。

4、其中SwrAllocSetOpts方法中有个参数libswresample.AvSampleFormat(1)，为什么取1，这里主要是选择采样表示方式的枚举，参考底层源码枚举，我发在下面。我这边因为音频是s16的，所以选择1。

enum AVSampleFormat {
    AV_SAMPLE_FMT_NONE = -1,
    AV_SAMPLE_FMT_U8,          ///< unsigned 8 bits
    AV_SAMPLE_FMT_S16,         ///< signed 16 bits
    AV_SAMPLE_FMT_S32,         ///< signed 32 bits
    AV_SAMPLE_FMT_FLT,         ///< float
    AV_SAMPLE_FMT_DBL,         ///< double
 
    AV_SAMPLE_FMT_U8P,         ///< unsigned 8 bits, planar
    AV_SAMPLE_FMT_S16P,        ///< signed 16 bits, planar
    AV_SAMPLE_FMT_S32P,        ///< signed 32 bits, planar
    AV_SAMPLE_FMT_FLTP,        ///< float, planar
    AV_SAMPLE_FMT_DBLP,        ///< double, planar
    AV_SAMPLE_FMT_S64,         ///< signed 64 bits
    AV_SAMPLE_FMT_S64P,        ///< signed 64 bits, planar
 
    AV_SAMPLE_FMT_NB           ///< Number of sample formats. DO NOT USE if linking dynamically
};

音频准备，输入音频为16k采样率音频。

(base) xxx@hu:~/GolandProjects/MediaRelay/data$ ffmpeg -i 1.wav 
ffmpeg version 4.2.7-0ubuntu0.1 Copyright (c) 2000-2022 the FFmpeg developers
  built with gcc 9 (Ubuntu 9.4.0-1ubuntu1~20.04.1)
  configuration: --prefix=/usr --extra-version=0ubuntu0.1 --toolchain=hardened --libdir=/usr/lib/x86_64-linux-gnu --incdir=/usr/include/x86_64-linux-gnu --arch=amd64 --enable-gpl --disable-stripping --enable-avresample --disable-filter=resample --enable-avisynth --enable-gnutls --enable-ladspa --enable-libaom --enable-libass --enable-libbluray --enable-libbs2b --enable-libcaca --enable-libcdio --enable-libcodec2 --enable-libflite --enable-libfontconfig --enable-libfreetype --enable-libfribidi --enable-libgme --enable-libgsm --enable-libjack --enable-libmp3lame --enable-libmysofa --enable-libopenjpeg --enable-libopenmpt --enable-libopus --enable-libpulse --enable-librsvg --enable-librubberband --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libspeex --enable-libssh --enable-libtheora --enable-libtwolame --enable-libvidstab --enable-libvorbis --enable-libvpx --enable-libwavpack --enable-libwebp --enable-libx265 --enable-libxml2 --enable-libxvid --enable-libzmq --enable-libzvbi --enable-lv2 --enable-omx --enable-openal --enable-opencl --enable-opengl --enable-sdl2 --enable-libdc1394 --enable-libdrm --enable-libiec61883 --enable-nvenc --enable-chromaprint --enable-frei0r --enable-libx264 --enable-shared
  libavutil      56. 31.100 / 56. 31.100
  libavcodec     58. 54.100 / 58. 54.100
  libavformat    58. 29.100 / 58. 29.100
  libavdevice    58.  8.100 / 58.  8.100
  libavfilter     7. 57.100 /  7. 57.100
  libavresample   4.  0.  0 /  4.  0.  0
  libswscale      5.  5.100 /  5.  5.100
  libswresample   3.  5.100 /  3.  5.100
  libpostproc    55.  5.100 / 55.  5.100
Guessed Channel Layout for Input Stream #0.0 : mono
Input #0, wav, from '1.wav':
  Metadata:
    date            : 2020-09-28
    encoder         : Lavf58.45.100
  Duration: 00:04:01.75, bitrate: 256 kb/s
    Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 16000 Hz, mono, s16, 256 kb/s

执行情况

input file format info -> AudioFormat:1,NumChannels:1,SampleRate:16000,ByteRate:32000,BlockAlign:2,BitsPerSample:16INFO : 2022/12/06 17:14:49.937547 csdn_wav_util.go:62: Create tempFile /tmp/2402235346.wav
11603961
 

最终音频

(base) xxx@hu:/tmp$ ffmpeg -i 2402235346.wav 
ffmpeg version 4.2.7-0ubuntu0.1 Copyright (c) 2000-2022 the FFmpeg developers
  built with gcc 9 (Ubuntu 9.4.0-1ubuntu1~20.04.1)
  configuration: --prefix=/usr --extra-version=0ubuntu0.1 --toolchain=hardened --libdir=/usr/lib/x86_64-linux-gnu --incdir=/usr/include/x86_64-linux-gnu --arch=amd64 --enable-gpl --disable-stripping --enable-avresample --disable-filter=resample --enable-avisynth --enable-gnutls --enable-ladspa --enable-libaom --enable-libass --enable-libbluray --enable-libbs2b --enable-libcaca --enable-libcdio --enable-libcodec2 --enable-libflite --enable-libfontconfig --enable-libfreetype --enable-libfribidi --enable-libgme --enable-libgsm --enable-libjack --enable-libmp3lame --enable-libmysofa --enable-libopenjpeg --enable-libopenmpt --enable-libopus --enable-libpulse --enable-librsvg --enable-librubberband --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libspeex --enable-libssh --enable-libtheora --enable-libtwolame --enable-libvidstab --enable-libvorbis --enable-libvpx --enable-libwavpack --enable-libwebp --enable-libx265 --enable-libxml2 --enable-libxvid --enable-libzmq --enable-libzvbi --enable-lv2 --enable-omx --enable-openal --enable-opencl --enable-opengl --enable-sdl2 --enable-libdc1394 --enable-libdrm --enable-libiec61883 --enable-nvenc --enable-chromaprint --enable-frei0r --enable-libx264 --enable-shared
  libavutil      56. 31.100 / 56. 31.100
  libavcodec     58. 54.100 / 58. 54.100
  libavformat    58. 29.100 / 58. 29.100
  libavdevice    58.  8.100 / 58.  8.100
  libavfilter     7. 57.100 /  7. 57.100
  libavresample   4.  0.  0 /  4.  0.  0
  libswscale      5.  5.100 /  5.  5.100
  libswresample   3.  5.100 /  3.  5.100
  libpostproc    55.  5.100 / 55.  5.100
Guessed Channel Layout for Input Stream #0.0 : mono
Input #0, wav, from '2402235346.wav':
  Duration: 00:04:01.75, bitrate: 768 kb/s
    Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 48000 Hz, mono, s16, 768 kb/s 

总结

其实在写代码过程中，有个让我特别头疼的东西，就是怎么把数组转为**uint。如果大家有兴趣可以研究一下ResampleByFFmpegApi2方法的转换逻辑，会学到不少东西。

分享：

        我们的疲劳往往不是由工作引起的，而是由于忧烦、挫折和不满等。——《人性的弱点》