如何在 rtsp 反向通道上发送音频数据?

How to send audio data on rtsp backchannel?

我正在使用支持音频输出(音频反向通道)的网络摄像机。我想要做的是通过提供的 RTSP URL 实时流式传输我的 PC 麦克风音频数据,这样在 PC 麦克风上所说的内容将在相机扬声器端听到。我已经阅读了关于 onvif 流媒体规范的信息,该规范告诉我,一旦我获得相机媒体的 RTSP url,我必须通过提供的 rtsp url 发送我的音频数据,以便在相机端进行音频输出。我的相机也支持 Onvif 配置文件 T。

到目前为止我尝试过的是-

  public static RtspClient rtspClient;
  public static IWaveIn sourceStream;

  private static void CallAudio()
    {
        string CameraIp = "192.168.1.69";
        string UserName = "admin";
        string Password = "admin123";
        var ClientMessageInspector = new ClientMessageInspector(UserName, Password);

        //Call Device Url and get Services.
        string DeviceServiceUrl = "http://" + CameraIp + "/onvif/device_service";
        var deviceClient = new DeviceClient("DeviceBinding", new EndpointAddress(DeviceServiceUrl));
        deviceClient.Endpoint.Behaviors.Add(ClientMessageInspector);
        var getServices = deviceClient.GetServices(false);                   
        
        //Call media2 getStreamingUri.
        string url = "http://" + CameraIp + "/onvif/media2_service";
        var Media2Client = new Media2Client("Media2Binding", new EndpointAddress(url));
        Media2Client.Endpoint.Behaviors.Add(ClientMessageInspector);
        var media2GetProfiles = Media2Client.GetProfiles(null, null);            
        var resp = Media2Client.GetAudioDecoderConfigurationOptions(null, null);
        var responseGetAudioStreamUri = Media2Client.GetStreamUri("tcp", profiles[0].token);  //This gets rtsp url of media from camera.

        rtspClient = new RtspClient(responseGetAudioStreamUri, UserName, Password);            
        sourceStream = new WaveInEvent();
        sourceStream.WaveFormat = new WaveFormat(64, 8, 1);   //8000 16
        sourceStream.DataAvailable += new EventHandler<WaveInEventArgs>(SourceStream_DataAvailable);
        sourceStream.StartRecording();

        Console.ReadKey();
    }

    //This method gets data from PC microphone and enocodes it into Mu-Law G711 and send to rtsp url.
    private static void SourceStream_DataAvailable(object sender, WaveInEventArgs e)
    {
        byte[] encoded = TwoWayAudio_Encode_MuLaw(e.Buffer, 0, e.BytesRecorded);
        rtspClient.SendData(encoded, encoded.Length, 3);
    }

    private static byte[] TwoWayAudio_Encode_MuLaw(byte[] data, int offset, int length)
    {
        byte[] encoded = new byte[length / 2];
        int outIndex = 0;
        for (int n = 0; n < length; n += 2)
        {
            encoded[outIndex++] = MuLawEncoder.LinearToMuLawSample(BitConverter.ToInt16(data, offset + n));
        }
        return encoded;
    }

我在项目中用于描述、设置和播放 rtsp URL 的 rtsp 客户端取自此 github 存储库 https://github.com/BogdanovKirill/RtspClientSharp

Rtspclient.cs

using Rtsp;
using Rtsp.Messages;
using Rtsp.Sdp;
using System;
using System.Collections.Generic;
using System.Diagnostics;
using System.IO;
using System.Security.Cryptography;
using System.Text;
using System.Text.RegularExpressions;

namespace Rtsp
{
   public class RtspClient
   {
    private RtspListener rtsp_client;
    private RtspTcpTransport tcp_socket;
    public string url;
    public bool canPlay = false;
    public string username;
    public string password;
    public ushort seqNo = 0;
    public event EventHandler<string> RtspError;
    public event EventHandler<byte[]> RtpDataReceived;
    public Stopwatch stopwatch { get; private set; }

    public RtspClient(string _url, string _username, string _password)
    {
        url = _url;
        username = _username;
        password = _password;

        var uri = new Uri(_url);

        tcp_socket = new RtspTcpTransport(uri.Host, 554); // 554);

        if (tcp_socket.Connected == false)
        {
            Console.WriteLine("Error - did not connect");
            return;
        }

        // Connect a RTSP Listener to the TCP Socket to send messages and listen for replies
        rtsp_client = new RtspListener(tcp_socket);

        rtsp_client.MessageReceived += Rtsp_client_MessageReceived;
        rtsp_client.DataReceived += DataReceived;
        rtsp_client.Start(); // start reading messages from the server
        rtsp_client.AutoReconnect = true;
        RtspRequest describe_message = new RtspRequestDescribe();
        describe_message.RtspUri = uri;            
        describe_message.AddHeader("Accept: application/sdp");
        describe_message.AddHeader("Require: www.onvif.org/ver20/backchannel");
        rtsp_client.SendMessage(describe_message);            

        stopwatch = new Stopwatch();
        stopwatch.Start();
    }

    private void DataReceived(object sender, RtspChunkEventArgs e)
    {
        int rtp_version = (e.Message.Data[0] >> 6);
        int rtp_padding = (e.Message.Data[0] >> 5) & 0x01;
        int rtp_extension = (e.Message.Data[0] >> 4) & 0x01;
        int rtp_csrc_count = (e.Message.Data[0] >> 0) & 0x0F;
        int rtp_marker = (e.Message.Data[1] >> 7) & 0x01;
        int rtp_payload_type = (e.Message.Data[1] >> 0) & 0x7F;
        uint rtp_sequence_number = ((uint)e.Message.Data[2] << 8) + (uint)(e.Message.Data[3]);
        uint rtp_timestamp = ((uint)e.Message.Data[4] << 24) + (uint)(e.Message.Data[5] << 16) + (uint)(e.Message.Data[6] << 8) + (uint)(e.Message.Data[7]);
        uint rtp_ssrc = ((uint)e.Message.Data[8] << 24) + (uint)(e.Message.Data[9] << 16) + (uint)(e.Message.Data[10] << 8) + (uint)(e.Message.Data[11]);

        int rtp_payload_start = 4 // V,P,M,SEQ
                            + 4 // time stamp
                            + 4 // ssrc
                            + (4 * rtp_csrc_count); // zero or more csrcs

        uint rtp_extension_id = 0;
        uint rtp_extension_size = 0;
        if (rtp_extension == 1)
        {
            rtp_extension_id = ((uint)e.Message.Data[rtp_payload_start + 0] << 8) + (uint)(e.Message.Data[rtp_payload_start + 1] << 0);
            rtp_extension_size = ((uint)e.Message.Data[rtp_payload_start + 2] << 8) + (uint)(e.Message.Data[rtp_payload_start + 3] << 0);
            rtp_payload_start += 4 + (int)rtp_extension_size;  // extension header and extension payload
        }

        Console.WriteLine("RTP Data"
                           + " V=" + rtp_version
                           + " P=" + rtp_padding
                           + " X=" + rtp_extension
                           + " CC=" + rtp_csrc_count
                           + " M=" + rtp_marker
                           + " PT=" + rtp_payload_type
                           + " Seq=" + rtp_sequence_number
                           + " Time=" + rtp_timestamp
                           + " SSRC=" + rtp_ssrc
                           + " Size=" + e.Message.Data.Length);


        // If rtp_marker is '1' then this is the final transmission for this packet.
        // If rtp_marker is '0' we need to accumulate data with the same timestamp

        // ToDo - Check Timestamp matches

        // Add to the tempoary_rtp List
        if (rtp_payload_type == 98 || rtp_payload_type == 0)
        {
            byte[] rtp_payload = new byte[e.Message.Data.Length - rtp_payload_start]; // payload with RTP header removed
            System.Array.Copy(e.Message.Data, rtp_payload_start, rtp_payload, 0, rtp_payload.Length); // copy payload
            RtpDataReceived?.Invoke(null, rtp_payload);
        }
    }

    public bool SendData(byte[] data, int count, int channel)
    {
        byte[] rtp_packet = new byte[12 + data.Length];
        int rtp_version = 2;
        int rtp_padding = 0;
        int rtp_extension = 0;
        int rtp_csrc_count = 0;
        int rtp_marker = 1; // set to 1 if the last NAL in the array
        //int rtp_payload_type = 98;
        int rtp_payload_type = 0;

        RTPPacketUtil.WriteHeader(rtp_packet, rtp_version, rtp_padding, rtp_extension, rtp_csrc_count, rtp_marker, rtp_payload_type);

        RTPPacketUtil.WriteSequenceNumber(rtp_packet, seqNo);
        seqNo++;
        RTPPacketUtil.WriteTS(rtp_packet, (uint)stopwatch.ElapsedMilliseconds);

        UInt32 empty_ssrc = 1293847657;
        RTPPacketUtil.WriteSSRC(rtp_packet, empty_ssrc);

        // Now append the raw NAL
        System.Array.Copy(data, 0, rtp_packet, 12, data.Length);

        if (canPlay)
        {
            rtsp_client.SendData(channel, rtp_packet);
            return true;
        }
        else return false;
    }

    private void Rtsp_client_MessageReceived(object sender, RtspChunkEventArgs e)
    {
        RtspResponse message = e.Message as RtspResponse;
        if (message.ReturnCode == 500)
        {
            RtspError?.Invoke(this, "Internal Server Error");
        }
        if (message.ReturnCode == 401)
        {
            Rtsp.Messages.RtspRequest msg = null;
            switch (message.OriginalRequest.Method)
            {
                case "DESCRIBE":
                    msg = new RtspRequestDescribe();
                    break;
                case "SETUP":
                    msg = new RtspRequestSetup();
                    break;
                default:
                    break;
            }
            msg.RtspUri = new Uri(url);
            var header = message.Headers["WWW-Authenticate"];
            var _realm = GrabHeaderVar("realm", header);
            var _nonce = GrabHeaderVar("nonce", header);
            var ha1 = CalculateMd5Hash(string.Format("{0}:{1}:{2}", username, _realm, password));
            var ha2 = CalculateMd5Hash(string.Format("{0}:{1}", message.OriginalRequest.Method, url));
            var digestResponse = CalculateMd5Hash(string.Format("{0}:{1}:{2}", ha1, _nonce, ha2));

            var digest = string.Format("Digest username=\"{0}\", realm=\"{1}\", nonce=\"{2}\", uri=\"{3}\", response=\"{4}\" ",
                username, _realm, _nonce, url, digestResponse);
            msg.AddHeader("Authorization: " + digest);
            msg.AddHeader("Accept: application/sdp");
            rtsp_client.SendMessage(msg);
            return;
        }
        Console.WriteLine("Received " + message.OriginalRequest.ToString());

        if (message.OriginalRequest != null && message.OriginalRequest is RtspRequestDescribe)
        {
            // Got a reply for DESCRIBE
            // Examine the SDP
            Console.Write(Encoding.UTF8.GetString(message.Data));

            SdpFile sdp_data;
            using (StreamReader sdp_stream = new StreamReader(new MemoryStream(message.Data)))
            {
                sdp_data = SdpFile.Read(sdp_stream);
            }

            // Process each 'Media' Attribute in the SDP.
            // If the attribute is for Video, then send a SETUP
            for (int x = 0; x < sdp_data.Medias.Count; x++)
            {
                if (sdp_data.Medias[x].MediaType == Rtsp.Sdp.Media.MediaTypes.audio || sdp_data.Medias[x].MediaType == Rtsp.Sdp.Media.MediaTypes.video)
                {
                    // seach the atributes for control, fmtp and rtpmap
                    String control = "";  // the "track" or "stream id"
                    String fmtp = ""; // holds SPS and PPS
                    String rtpmap = ""; // holds the Payload format, 96 is often used with H264
                    foreach (Rtsp.Sdp.Attribut attrib in sdp_data.Medias[x].Attributs)
                    {
                        if (attrib.Key.Equals("control")) control = attrib.Value;
                        if (attrib.Key.Equals("fmtp")) fmtp = attrib.Value;
                        if (attrib.Key.Equals("rtpmap")) rtpmap = attrib.Value;
                    }

                    // Get the Payload format number for the Video Stream
                    String[] split_rtpmap = rtpmap.Split(' ');
                    var video_payload = 0;
                    bool result = Int32.TryParse(split_rtpmap[0], out video_payload);
                                      
                    // Send SETUP for the Video Stream
                    // using Interleaved mode (RTP frames over the RTSP socket)
                    Rtsp.Messages.RtspRequest setup_message = new Rtsp.Messages.RtspRequestSetup();
                    setup_message.RtspUri = new Uri(url + "/" + control);                        
                    //setup_message.AddHeader("Transport: RTP/AVP/TCP;interleaved=0");                       
                    setup_message.AddHeader("Require: www.onvif.org/ver20/backchannel");                                              
                    rtsp_client.SendMessage(setup_message);                                            
                }
            }
        }

        if (message.OriginalRequest != null && message.OriginalRequest is RtspRequestSetup)
        {
            // Got Reply to SETUP
            Console.WriteLine("Got reply from Setup. Session is " + message.Session);

            String session = message.Session; // Session value used with Play, Pause, Teardown

            // Send PLAY
            RtspRequest play_message = new RtspRequestPlay();
            play_message.RtspUri = new Uri(url);               
            play_message.Session = session;
            play_message.AddHeader("Require: www.onvif.org/ver20/backchannel");
            rtsp_client.SendMessage(play_message);
        }

        if (message.OriginalRequest != null && message.OriginalRequest is RtspRequestPlay)
        {
            // Got Reply to PLAY
            Console.WriteLine("Got reply from Play  " + message.Command);
            canPlay = true;

        }
    }

    private static string GrabHeaderVar(string varName, string header)
    {
        var regHeader = new Regex(string.Format(@"{0}=""([^""]*)""", varName));
        var matchHeader = regHeader.Match(header);
        if (matchHeader.Success)
            return matchHeader.Groups[1].Value;
        throw new ApplicationException(string.Format("Header {0} not found", varName));
    }

    private static string CalculateMd5Hash(string input)
    {
        var inputBytes = Encoding.ASCII.GetBytes(input);
        var hash = MD5.Create().ComputeHash(inputBytes);
        var sb = new StringBuilder();
        foreach (var b in hash)
            sb.Append(b.ToString("x2"));
        return sb.ToString();
    }

    public void Dispose()
    {
        rtsp_client.Stop();
        rtsp_client.Dispose();
    }
}

}

所以在调用任何 rtsp 方法之前,我添加了 Rtsp 要求:www.onvif.org/ver20/backchannel 这对于检查相机是否支持 AudioBack 通道很重要。

调用 Describe、Setup 和 play 后得到的输出正常。

Received Rtsp.Messages.RtspRequestDescribe
v=0
o=- 0 0 IN IP4 192.168.1.69
s=LIVE VIEW
c=IN IP4 0.0.0.0
t=0 0
a=control:rtsp://192.168.1.69/rtsp_tunnel?p=0&h26x=4&aon=1&aud=0
m=video 0 RTP/AVP 35
a=rtpmap:35 H264/90000
a=control:rtsp://192.168.1.69/rtsp_tunnel?p=0&h26x=4&aon=1&aud=0&stream=video
a=recvonly
a=fmtp:35 packetization-mode=1;profile-level-id=4d0029;sprop-parameter- 
sets=Z00AKZpkA8ARPy4C1BQEFAg=,aO48gA==
m=audio 0 RTP/AVP 96
a=rtpmap:96 mpeg4-generic/16000/1
a=fmtp:96 streamtype=5; profile-level-id=5; mode=AAC-hbr; config=1408; SizeLength=13; IndexLength=3; 
IndexDeltaLength=3
a=control:rtsp://192.168.1.69/rtsp_tunnel?p=0&h26x=4&aon=1&aud=0&stream=audio
a=recvonly
m=audio 0 RTP/AVP 0
a=rtpmap:0 PCMU/8000/1
a=control:rtsp://192.168.1.69/rtsp_tunnel?p=0&h26x=4&aon=1&aud=0&stream=backchannel
a=sendonly
Received Rtsp.Messages.RtspRequestSetup
Got reply from Setup. Session is 12346e9856840dc
Received Rtsp.Messages.RtspRequestSetup
Got reply from Setup. Session is 12346e9856840dc
Received Rtsp.Messages.RtspRequestSetup
Got reply from Setup. Session is 12346e9856840dc
Received Rtsp.Messages.RtspRequestPlay
Got reply from Play  RTSP/1.0 200 OK
Received Rtsp.Messages.RtspRequestPlay
Got reply from Play  RTSP/1.0 200 OK
Received Rtsp.Messages.RtspRequestPlay
Got reply from Play  RTSP/1.0 200 OK

从 Play 方法获得响应后,我开始使用 Rtsp 客户端中的发送方法发送我的编码数据。

但是我在摄像头端听不到音频。我的问题很简单-

  1. 是否可以通过 RTSP 发送音频数据URL。
  2. 我调用方法的方式有什么问题吗?请指出。
  3. 有没有简单的方法(或example/tutorial)显示如何完成我的任务,请提供给。

请不要介意我是 Rtsp 的新手。提前致谢。如果问题有任何问题,请告诉我,我会对其进行编辑以便清楚地理解。


我发现我做错了什么。其实上面提到的步骤很好,相机返回好的,我做错的是上面的代码:

sourceStream.WaveFormat = new WaveFormat(64, 8, 1);   //8000 16

64 和 8 参数应该是:

sourceStream.WaveFormat = new WaveFormat(8000, 16, 1);   //8000 16

都是语音采样率导致语音发送听不见。谢谢!