OLD | NEW |
---|---|
1 /* | 1 /* |
2 * Copyright (c) 2017 The WebRTC project authors. All Rights Reserved. | 2 * Copyright (c) 2017 The WebRTC project authors. All Rights Reserved. |
3 * | 3 * |
4 * Use of this source code is governed by a BSD-style license | 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 | 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 | 6 * tree. An additional intellectual property rights grant can be found |
7 * in the file PATENTS. All contributing project authors may | 7 * in the file PATENTS. All contributing project authors may |
8 * be found in the AUTHORS file in the root of the source tree. | 8 * be found in the AUTHORS file in the root of the source tree. |
9 */ | 9 */ |
10 | 10 |
11 #include "webrtc/modules/audio_processing/test/conversational_speech/multiend_ca ll.h" | 11 #include "webrtc/modules/audio_processing/test/conversational_speech/multiend_ca ll.h" |
12 | 12 |
13 #include <algorithm> | 13 #include <algorithm> |
14 #include <iterator> | 14 #include <iterator> |
15 | 15 |
16 #include "webrtc/base/logging.h" | 16 #include "webrtc/base/logging.h" |
17 #include "webrtc/base/pathutils.h" | 17 #include "webrtc/base/pathutils.h" |
18 | 18 |
19 namespace webrtc { | 19 namespace webrtc { |
20 namespace test { | 20 namespace test { |
21 namespace conversational_speech { | 21 namespace conversational_speech { |
22 | 22 |
23 MultiEndCall::MultiEndCall( | 23 MultiEndCall::MultiEndCall( |
24 rtc::ArrayView<const Turn> timing, const std::string& audiotracks_path, | 24 rtc::ArrayView<const Turn> timing, const std::string& audiotracks_path, |
25 std::unique_ptr<WavReaderAbstractFactory> wavreader_abstract_factory) | 25 std::unique_ptr<WavReaderAbstractFactory> wavreader_abstract_factory) |
26 : timing_(timing), audiotracks_path_(audiotracks_path), | 26 : timing_(timing), audiotracks_path_(audiotracks_path), |
27 wavreader_abstract_factory_(std::move(wavreader_abstract_factory)) { | 27 wavreader_abstract_factory_(std::move(wavreader_abstract_factory)), |
28 valid_(false) { | |
28 FindSpeakerNames(); | 29 FindSpeakerNames(); |
29 CreateAudioTrackReaders(); | 30 if (CreateAudioTrackReaders()) |
30 valid_ = CheckTiming(); | 31 valid_ = CheckTiming(); |
31 } | 32 } |
32 | 33 |
33 MultiEndCall::~MultiEndCall() = default; | 34 MultiEndCall::~MultiEndCall() = default; |
34 | 35 |
35 const std::set<std::string>& MultiEndCall::speaker_names() const { | 36 const std::set<std::string>& MultiEndCall::speaker_names() const { |
36 return speaker_names_; | 37 return speaker_names_; |
37 } | 38 } |
38 | 39 |
39 const std::map<std::string, std::unique_ptr<WavReaderInterface>>& | 40 const std::map<std::string, std::unique_ptr<WavReaderInterface>>& |
40 MultiEndCall::audiotrack_readers() const { | 41 MultiEndCall::audiotrack_readers() const { |
41 return audiotrack_readers_; | 42 return audiotrack_readers_; |
42 } | 43 } |
43 | 44 |
44 bool MultiEndCall::valid() const { | 45 bool MultiEndCall::valid() const { |
45 return valid_; | 46 return valid_; |
46 } | 47 } |
47 | 48 |
49 int MultiEndCall::sample_rate() const { | |
50 return sample_rate_; | |
51 } | |
52 | |
48 size_t MultiEndCall::total_duration_samples() const { | 53 size_t MultiEndCall::total_duration_samples() const { |
49 return total_duration_samples_; | 54 return total_duration_samples_; |
50 } | 55 } |
51 | 56 |
52 const std::vector<MultiEndCall::SpeakingTurn>& MultiEndCall::speaking_turns() | 57 const std::vector<MultiEndCall::SpeakingTurn>& MultiEndCall::speaking_turns() |
53 const { | 58 const { |
54 return speaking_turns_; | 59 return speaking_turns_; |
55 } | 60 } |
56 | 61 |
57 void MultiEndCall::FindSpeakerNames() { | 62 void MultiEndCall::FindSpeakerNames() { |
58 RTC_DCHECK(speaker_names_.empty()); | 63 RTC_DCHECK(speaker_names_.empty()); |
59 for (const Turn& turn : timing_) { | 64 for (const Turn& turn : timing_) { |
60 speaker_names_.emplace(turn.speaker_name); | 65 speaker_names_.emplace(turn.speaker_name); |
61 } | 66 } |
62 } | 67 } |
63 | 68 |
64 void MultiEndCall::CreateAudioTrackReaders() { | 69 bool MultiEndCall::CreateAudioTrackReaders() { |
65 RTC_DCHECK(audiotrack_readers_.empty()); | 70 RTC_DCHECK(audiotrack_readers_.empty()); |
71 sample_rate_ = 0; // Sample rate will be set when reading the first track. | |
66 for (const Turn& turn : timing_) { | 72 for (const Turn& turn : timing_) { |
67 auto it = audiotrack_readers_.find(turn.audiotrack_file_name); | 73 auto it = audiotrack_readers_.find(turn.audiotrack_file_name); |
68 if (it != audiotrack_readers_.end()) | 74 if (it != audiotrack_readers_.end()) |
minyue-webrtc
2017/05/16 15:05:16
If this is supposed to be incremental, why doesn't
AleBzk
2017/05/17 12:49:37
Here I check if the audio track file has already b
| |
69 continue; | 75 continue; |
70 | 76 |
71 // Instance Pathname to retrieve the full path to the audiotrack file. | 77 // Instance Pathname to retrieve the full path to the audiotrack file. |
72 const rtc::Pathname audiotrack_file_path( | 78 const rtc::Pathname audiotrack_file_path( |
73 audiotracks_path_, turn.audiotrack_file_name); | 79 audiotracks_path_, turn.audiotrack_file_name); |
74 | 80 |
75 // Map the audiotrack file name to a new instance of WavReaderInterface. | 81 // Map the audiotrack file name to a new instance of WavReaderInterface. |
76 std::unique_ptr<WavReaderInterface> wavreader = | 82 std::unique_ptr<WavReaderInterface> wavreader = |
77 wavreader_abstract_factory_->Create(audiotrack_file_path.pathname()); | 83 wavreader_abstract_factory_->Create(audiotrack_file_path.pathname()); |
84 | |
85 if (sample_rate_ == 0) { | |
86 sample_rate_ = wavreader->SampleRate(); | |
87 } else if (sample_rate_ != wavreader->SampleRate()) { | |
88 LOG(LS_ERROR) << "all the audio tracks should have the same sample rate"; | |
minyue-webrtc
2017/05/16 15:05:16
"A"ll and add full stop at the end of the sentence
AleBzk
2017/05/17 12:49:37
Done.
| |
89 return false; | |
90 } | |
91 | |
92 if (wavreader->NumChannels() != 1) { | |
93 LOG(LS_ERROR) << "only mono audio tracks supported"; | |
minyue-webrtc
2017/05/16 15:05:16
"O"nly and full stop.
AleBzk
2017/05/17 12:49:37
Done.
| |
94 return false; | |
95 } | |
96 | |
78 audiotrack_readers_.emplace( | 97 audiotrack_readers_.emplace( |
minyue-webrtc
2017/05/16 15:05:16
Is this emplace allowed? try a few trybots.
AleBzk
2017/05/17 12:49:37
No need to try. This line was landed in a previous
minyue-webrtc
2017/05/18 13:35:13
Acknowledged.
| |
79 turn.audiotrack_file_name, std::move(wavreader)); | 98 turn.audiotrack_file_name, std::move(wavreader)); |
80 } | 99 } |
100 | |
101 return true; | |
81 } | 102 } |
82 | 103 |
83 bool MultiEndCall::CheckTiming() { | 104 bool MultiEndCall::CheckTiming() { |
84 struct Interval { | 105 struct Interval { |
85 size_t begin; | 106 size_t begin; |
86 size_t end; | 107 size_t end; |
87 }; | 108 }; |
88 size_t number_of_turns = timing_.size(); | 109 size_t number_of_turns = timing_.size(); |
89 auto millisecond_to_samples = [](int ms, int sr) -> int { | 110 auto millisecond_to_samples = [](int ms, int sr) -> int { |
90 // Truncation may happen if the sampling rate is not an integer multiple | 111 // Truncation may happen if the sampling rate is not an integer multiple |
(...skipping 95 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... | |
186 return false; | 207 return false; |
187 } | 208 } |
188 } | 209 } |
189 | 210 |
190 return true; | 211 return true; |
191 } | 212 } |
192 | 213 |
193 } // namespace conversational_speech | 214 } // namespace conversational_speech |
194 } // namespace test | 215 } // namespace test |
195 } // namespace webrtc | 216 } // namespace webrtc |
OLD | NEW |