Home | History | Annotate | Download | only in media 
      1 /*
      2  * libjingle
      3  * Copyright 2011 Google Inc.
      4  *
      5  * Redistribution and use in source and binary forms, with or without
      6  * modification, are permitted provided that the following conditions are met:
      7  *
      8  *  1. Redistributions of source code must retain the above copyright notice,
      9  *     this list of conditions and the following disclaimer.
     10  *  2. Redistributions in binary form must reproduce the above copyright notice,
     11  *     this list of conditions and the following disclaimer in the documentation
     12  *     and/or other materials provided with the distribution.
     13  *  3. The name of the author may not be used to endorse or promote products
     14  *     derived from this software without specific prior written permission.
     15  *
     16  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
     17  * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
     18  * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
     19  * EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
     20  * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
     21  * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
     22  * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
     23  * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
     24  * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
     25  * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
     26  */
     27 
     28 // CurrentSpeakerMonitor monitors the audio levels for a session and determines
     29 // which participant is currently speaking.
     30 
     31 #ifndef TALK_SESSION_MEDIA_CURRENTSPEAKERMONITOR_H_
     32 #define TALK_SESSION_MEDIA_CURRENTSPEAKERMONITOR_H_
     33 
     34 #include <map>
     35 
     36 #include "webrtc/base/basictypes.h"
     37 #include "webrtc/base/sigslot.h"
     38 
     39 namespace cricket {
     40 
     41 struct AudioInfo;
     42 struct MediaStreams;
     43 
     44 class AudioSourceContext {
     45  public:
     46   sigslot::signal2<AudioSourceContext*, const cricket::AudioInfo&>
     47       SignalAudioMonitor;
     48   sigslot::signal1<AudioSourceContext*> SignalMediaStreamsReset;
     49   sigslot::signal3<AudioSourceContext*,
     50                    const cricket::MediaStreams&,
     51                    const cricket::MediaStreams&> SignalMediaStreamsUpdate;
     52 };
     53 
     54 // CurrentSpeakerMonitor can be used to monitor the audio-levels from
     55 // many audio-sources and report on changes in the loudest audio-source.
     56 // Its a generic type and relies on an AudioSourceContext which is aware of
     57 // the audio-sources. AudioSourceContext needs to provide two signals namely
     58 // SignalAudioInfoMonitor - provides audio info of the all current speakers.
     59 // SignalMediaSourcesUpdated - provides updates when a speaker leaves or joins.
     60 // Note that the AudioSourceContext's audio monitor must be started
     61 // before this is started.
     62 // It's recommended that the audio monitor be started with a 100 ms period.
     63 class CurrentSpeakerMonitor : public sigslot::has_slots<> {
     64  public:
     65   CurrentSpeakerMonitor(AudioSourceContext* audio_source_context);
     66   ~CurrentSpeakerMonitor();
     67 
     68   void Start();
     69   void Stop();
     70 
     71   // Used by tests.  Note that the actual minimum time between switches
     72   // enforced by the monitor will be the given value plus or minus the
     73   // resolution of the system clock.
     74   void set_min_time_between_switches(uint32_t min_time_between_switches);
     75 
     76   // This is fired when the current speaker changes, and provides his audio
     77   // SSRC.  This only fires after the audio monitor on the underlying
     78   // AudioSourceContext has been started.
     79   sigslot::signal2<CurrentSpeakerMonitor*, uint32_t> SignalUpdate;
     80 
     81  private:
     82   void OnAudioMonitor(AudioSourceContext* audio_source_context,
     83                       const AudioInfo& info);
     84   void OnMediaStreamsUpdate(AudioSourceContext* audio_source_context,
     85                             const MediaStreams& added,
     86                             const MediaStreams& removed);
     87   void OnMediaStreamsReset(AudioSourceContext* audio_source_context);
     88 
     89   // These are states that a participant will pass through so that we gradually
     90   // recognize that they have started and stopped speaking.  This avoids
     91   // "twitchiness".
     92   enum SpeakingState {
     93     SS_NOT_SPEAKING,
     94     SS_MIGHT_BE_SPEAKING,
     95     SS_SPEAKING,
     96     SS_WAS_SPEAKING_RECENTLY1,
     97     SS_WAS_SPEAKING_RECENTLY2
     98   };
     99 
    100   bool started_;
    101   AudioSourceContext* audio_source_context_;
    102   std::map<uint32_t, SpeakingState> ssrc_to_speaking_state_map_;
    103   uint32_t current_speaker_ssrc_;
    104   // To prevent overswitching, switching is disabled for some time after a
    105   // switch is made.  This gives us the earliest time a switch is permitted.
    106   uint32_t earliest_permitted_switch_time_;
    107   uint32_t min_time_between_switches_;
    108 };
    109 
    110 }
    111 
    112 #endif  // TALK_SESSION_MEDIA_CURRENTSPEAKERMONITOR_H_
    113