以RTP形式从设备发送NAudio/Opus编码音频
本文关键字:NAudio Opus 音频 编码 RTP | 更新日期: 2023-09-27 18:28:17
首先,我道歉。很长一段时间前,我曾对VB5进行过修补,多年来一直不在程序员的岗位上——我仍在重新学习基础知识,最近开始学习C#/.NET。我也是这个网站的新手,请你耐心等待和指导。关于我的背景故事够多了。
使用Opus的这个包装器,我将包装器项目添加到了自己的解决方案中,并且NAudio我相信我已经设置好了,可以从我的设备(声卡)中主动获取音频,并利用示例编码器代码将编码的音频获取到_playBuffer中。
我的下一个任务是从中获取编码数据,并使用RDP发送,这样它就可以在另一台机器上的客户端应用程序中进行解码,在那里它将被解码并在他们的声音设备中播放。
我对_playBuffer中的数据已准备好进行编码数据的理解是否正确?还是需要对RTP数据包进行不同的拆分?(我在这里看到了一个uLAW的例子,但不确定我是否能适应我的需求。由于下载的源代码是用德语注释的,但我几乎不会把英语作为第一语言说和写,即使是这些也没有太大帮助。)
(我是否使用了正确的术语?)到目前为止,你看到的股票代码通过WaveOut将_playBuffer数据放回原处,就像他的例子一样——我在这里忽略了这一点,留下来解释我(可能缺乏)的理解。(如果它是"可播放的",它就是"可发送的"。)
另一个问题是,我的意图是在互联网上点对点地多播流——尽管我不确定多播是否是我想要的。
using System;
using System.Collections.Generic;
using System.ComponentModel;
using System.Data;
using System.Drawing;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
using System.Windows.Forms;
using NAudio;
using NAudio.CoreAudioApi;
using NAudio.Wave;
using FragLabs.Audio.Codecs;
namespace VUmeterappStereo
{
public partial class Form1 : Form
{private void Form1_Load(object sender, EventArgs e)
{
for (int i = 0; i < WaveIn.DeviceCount; i++)
{
comboBox1.Items.Add(WaveIn.GetCapabilities(i).ProductName);
}
if (WaveIn.DeviceCount > 0)
comboBox1.SelectedIndex = 0;
for (int i = 0; i < WaveOut.DeviceCount; i++)
{
comboBox2.Items.Add(WaveOut.GetCapabilities(i).ProductName);
}
if (WaveOut.DeviceCount > 0)
comboBox2.SelectedIndex = 0;
}
private void button1_Click(object sender, EventArgs e)
{
button2.Enabled = true;
button1.Enabled = false;
StartEncoding();
}
private void button2_Click(object sender, EventArgs e)
{
button1.Enabled = true;
button2.Enabled = false;
StopEncoding();
}
WaveIn _waveIn;
WaveOut _waveOut;
BufferedWaveProvider _playBuffer;
OpusEncoder _encoder;
OpusDecoder _decoder;
int _segmentFrames;
int _bytesPerSegment;
ulong _bytesSent;
DateTime _startTime;
Timer _timer = null;
void StartEncoding()
{
_startTime = DateTime.Now;
_bytesSent = 0;
_segmentFrames = 960;
_encoder = OpusEncoder.Create(48000, 1, FragLabs.Audio.Codecs.Opus.Application.Voip);
_encoder.Bitrate = 8192;
_decoder = OpusDecoder.Create(48000, 1);
_bytesPerSegment = _encoder.FrameByteCount(_segmentFrames);
_waveIn = new WaveIn(WaveCallbackInfo.FunctionCallback());
_waveIn.BufferMilliseconds = 50;
_waveIn.DeviceNumber = comboBox1.SelectedIndex;
_waveIn.DataAvailable += _waveIn_DataAvailable;
_waveIn.WaveFormat = new WaveFormat(48000, 16, 1);
_playBuffer = new BufferedWaveProvider(new WaveFormat(48000, 16, 1));
_waveOut = new WaveOut(WaveCallbackInfo.FunctionCallback());
_waveOut.DeviceNumber = comboBox2.SelectedIndex;
_waveOut.Init(_playBuffer);
_waveOut.Play();
_waveIn.StartRecording();
if (_timer == null)
{
_timer = new Timer();
_timer.Interval = 1000;
_timer.Tick += _timer_Tick;
}
_timer.Start();
}
void _timer_Tick(object sender, EventArgs e)
{
var timeDiff = DateTime.Now - _startTime;
var bytesPerSecond = _bytesSent / timeDiff.TotalSeconds;
Console.WriteLine("{0} Bps", bytesPerSecond);
}
byte[] _notEncodedBuffer = new byte[0];
void _waveIn_DataAvailable(object sender, WaveInEventArgs e)
{
byte[] soundBuffer = new byte[e.BytesRecorded + _notEncodedBuffer.Length];
for (int i = 0; i < _notEncodedBuffer.Length; i++)
soundBuffer[i] = _notEncodedBuffer[i];
for (int i = 0; i < e.BytesRecorded; i++)
soundBuffer[i + _notEncodedBuffer.Length] = e.Buffer[i];
int byteCap = _bytesPerSegment;
int segmentCount = (int)Math.Floor((decimal)soundBuffer.Length / byteCap);
int segmentsEnd = segmentCount * byteCap;
int notEncodedCount = soundBuffer.Length - segmentsEnd;
_notEncodedBuffer = new byte[notEncodedCount];
for (int i = 0; i < notEncodedCount; i++)
{
_notEncodedBuffer[i] = soundBuffer[segmentsEnd + i];
}
for (int i = 0; i < segmentCount; i++)
{
byte[] segment = new byte[byteCap];
for (int j = 0; j < segment.Length; j++)
segment[j] = soundBuffer[(i * byteCap) + j];
int len;
byte[] buff = _encoder.Encode(segment, segment.Length, out len);
_bytesSent += (ulong)len;
buff = _decoder.Decode(buff, len, out len);
_playBuffer.AddSamples(buff, 0, len);
}
}
void StopEncoding()
{
_timer.Stop();
_waveIn.StopRecording();
_waveIn.Dispose();
_waveIn = null;
_waveOut.Stop();
_waveOut.Dispose();
_waveOut = null;
_playBuffer = null;
_encoder.Dispose();
_encoder = null;
_decoder.Dispose();
_decoder = null;
}
private void timer1_Tick(object sender, EventArgs e)
{
MMDeviceEnumerator de = new MMDeviceEnumerator();
MMDevice device = de.GetDefaultAudioEndpoint(DataFlow.Render, Role.Multimedia);
//float volume = (float)device.AudioMeterInformation.MasterPeakValue * 100;
float volLeft = (float)device.AudioMeterInformation.PeakValues[0] * 100;
float volRight = (float)device.AudioMeterInformation.PeakValues[1] * 100;
progressBar1.Value = (int)volLeft;
progressBar2.Value = (int)volRight;
}
private void timer2_Tick(object sender, EventArgs e)
{
}
}
}
感谢您的任何贡献,帮助我了解如何通过RTP流获取数据。
哦,是的,这首先是从我从一个教程示例中重新创建VU仪表的修补开始的——因此是名称空间名称和额外的代码,它确实起作用。
代码示例对音频进行编码而非解码。您需要将Buff中包含的字节发送到网络。
上面示例中的这段代码正在接收来自声卡的音频。
byte[] _notEncodedBuffer = new byte[0];
void _waveIn_DataAvailable(object sender, WaveInEventArgs e)
{
byte[] soundBuffer = new byte[e.BytesRecorded + _notEncodedBuffer.Length];
for (int i = 0; i < _notEncodedBuffer.Length; i++)
soundBuffer[i] = _notEncodedBuffer[i];
for (int i = 0; i < e.BytesRecorded; i++)
soundBuffer[i + _notEncodedBuffer.Length] = e.Buffer[i];
int byteCap = _bytesPerSegment;
int segmentCount = (int)Math.Floor((decimal)soundBuffer.Length / byteCap);
int segmentsEnd = segmentCount * byteCap;
int notEncodedCount = soundBuffer.Length - segmentsEnd;
_notEncodedBuffer = new byte[notEncodedCount];
for (int i = 0; i < notEncodedCount; i++)
{
_notEncodedBuffer[i] = soundBuffer[segmentsEnd + i];
}
for (int i = 0; i < segmentCount; i++)
{
byte[] segment = new byte[byteCap];
for (int j = 0; j < segment.Length; j++)
segment[j] = soundBuffer[(i * byteCap) + j];
int len;
byte[] buff = _encoder.Encode(segment, segment.Length, out len);
_bytesSent += (ulong)len;
buff = _decoder.Decode(buff, len, out len);
_playBuffer.AddSamples(buff, 0, len);
}
}
在这条线上
byte[]buff=_encoder.Encode(segment,segment.Length,out len);
正是在这一点上,你创建了你的RTP数据包
https://www.rfc-editor.org/rfc/rfc3550
然后使用C#在网络上发送
通常作为UDP
在C#中发送UDP数据包
在从RTP数据包中提取Buff之后,剩余的代码属于接收应用程序。
buff = _decoder.Decode(buff, len, out len);
_playBuffer.AddSamples(buff, 0, len);