1 /*
2  *  Copyright (c) 2019 The WebRTC project authors. All Rights Reserved.
3  *
4  *  Use of this source code is governed by a BSD-style license
5  *  that can be found in the LICENSE file in the root of the source
6  *  tree. An additional intellectual property rights grant can be found
7  *  in the file PATENTS.  All contributing project authors may
8  *  be found in the AUTHORS file in the root of the source tree.
9  */
10 
11 #ifndef MODULES_AUDIO_PROCESSING_VOICE_DETECTION_H_
12 #define MODULES_AUDIO_PROCESSING_VOICE_DETECTION_H_
13 
14 #include <stddef.h>
15 
16 #include <memory>
17 
18 #include "modules/audio_processing/include/audio_processing.h"
19 
20 namespace webrtc {
21 
22 class AudioBuffer;
23 
24 // The voice activity detection (VAD) component analyzes the stream to
25 // determine if voice is present.
26 class VoiceDetection {
27  public:
28   // Specifies the likelihood that a frame will be declared to contain voice.
29   // A higher value makes it more likely that speech will not be clipped, at
30   // the expense of more noise being detected as voice.
31   enum Likelihood {
32     kVeryLowLikelihood,
33     kLowLikelihood,
34     kModerateLikelihood,
35     kHighLikelihood
36   };
37 
38   VoiceDetection(int sample_rate_hz, Likelihood likelihood);
39   ~VoiceDetection();
40 
41   VoiceDetection(VoiceDetection&) = delete;
42   VoiceDetection& operator=(VoiceDetection&) = delete;
43 
44   // Returns true if voice is detected in the current frame.
45   bool ProcessCaptureAudio(AudioBuffer* audio);
46 
likelihood()47   Likelihood likelihood() const { return likelihood_; }
48 
49  private:
50   class Vad;
51 
52   int sample_rate_hz_;
53   size_t frame_size_samples_;
54   Likelihood likelihood_;
55   std::unique_ptr<Vad> vad_;
56 };
57 }  // namespace webrtc
58 
59 #endif  // MODULES_AUDIO_PROCESSING_VOICE_DETECTION_H_
60