OLD | NEW |
---|---|
1 /* | 1 /* |
2 * Copyright (c) 2012 The WebRTC project authors. All Rights Reserved. | 2 * Copyright (c) 2012 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 #ifndef WEBRTC_MODULES_AUDIO_CODING_NETEQ_INCLUDE_AUDIO_DECODER_H_ | 11 #ifndef WEBRTC_MODULES_AUDIO_CODING_NETEQ_INCLUDE_AUDIO_DECODER_H_ |
12 #define WEBRTC_MODULES_AUDIO_CODING_NETEQ_INCLUDE_AUDIO_DECODER_H_ | 12 #define WEBRTC_MODULES_AUDIO_CODING_NETEQ_INCLUDE_AUDIO_DECODER_H_ |
13 | 13 |
14 #include <stdlib.h> // NULL | 14 #include <stdlib.h> // NULL |
15 | 15 |
16 #include "webrtc/base/array_view.h" | |
17 #include "webrtc/base/buffer.h" | |
16 #include "webrtc/base/constructormagic.h" | 18 #include "webrtc/base/constructormagic.h" |
19 #include "webrtc/base/optional.h" | |
17 #include "webrtc/typedefs.h" | 20 #include "webrtc/typedefs.h" |
18 | 21 |
19 namespace webrtc { | 22 namespace webrtc { |
20 | 23 |
21 // This is the interface class for decoders in NetEQ. Each codec type will have | 24 // This is the interface class for decoders in NetEQ. Each codec type will have |
22 // and implementation of this class. | 25 // and implementation of this class. |
23 class AudioDecoder { | 26 class AudioDecoder { |
24 public: | 27 public: |
25 enum SpeechType { | 28 enum SpeechType { |
26 kSpeech = 1, | 29 kSpeech = 1, |
27 kComfortNoise = 2 | 30 kComfortNoise = 2 |
28 }; | 31 }; |
29 | 32 |
30 // Used by PacketDuration below. Save the value -1 for errors. | 33 // Used by PacketDuration below. Save the value -1 for errors. |
31 enum { kNotImplemented = -2 }; | 34 enum { kNotImplemented = -2 }; |
32 | 35 |
33 AudioDecoder() = default; | 36 AudioDecoder() = default; |
34 virtual ~AudioDecoder() = default; | 37 virtual ~AudioDecoder() = default; |
35 | 38 |
39 class EncodedAudioFrame { | |
40 public: | |
41 struct DecodeResult { | |
42 size_t num_decoded_samples; | |
43 SpeechType speech_type; | |
44 }; | |
45 | |
46 virtual ~EncodedAudioFrame() = default; | |
47 | |
48 // Returns the duration in samples-per-channel of this audio frame. | |
49 // If no duration can be ascertained, returns zero. | |
50 virtual size_t Duration() const = 0; | |
51 | |
52 // Decodes this frame of audio and writes the result in |decoded|. | |
53 // |decoded| will be large enough for 120 ms of audio at the decoder's | |
54 // sample rate. On success, returns an rtc::Optional containing the total | |
kwiberg-webrtc
2016/09/16 00:14:07
Might it be better to simply state that |decoded|
ossu
2016/09/16 11:24:16
I've incorporated the phrasing changes suggested i
| |
55 // number of samples across all channels, as well as whether the decoder | |
56 // produced comfort noise or speech. On failure, returns an empty | |
57 // rtc::Optional. Decode must be called at most once per frame object. | |
kwiberg-webrtc
2016/09/16 00:14:07
My sometimes reliable sense of English suggests s/
ossu
2016/09/16 11:24:16
I believe you're right.
| |
58 virtual rtc::Optional<DecodeResult> Decode( | |
59 rtc::ArrayView<int16_t> decoded) const = 0; | |
60 }; | |
61 | |
62 struct ParseResult { | |
63 ParseResult(); | |
64 ParseResult(uint32_t timestamp, | |
65 bool primary, | |
66 std::unique_ptr<EncodedAudioFrame> frame); | |
67 ParseResult(ParseResult&& b); | |
68 ~ParseResult(); | |
69 | |
70 ParseResult& operator=(ParseResult&& b); | |
71 | |
72 // The timestamp of the frame is in samples per channel. | |
73 uint32_t timestamp; | |
74 bool primary; | |
75 std::unique_ptr<EncodedAudioFrame> frame; | |
76 }; | |
77 | |
78 // Let the decoder parse this payload and prepare zero or more decodable | |
79 // frames. Each frame must be at most 120 ms long and should never be shorter | |
80 // than 10 ms. The caller must ensure that the AudioDecoder object outlives | |
kwiberg-webrtc
2016/09/16 00:14:07
This use of "must" and "should" gives the sense th
hlundin-webrtc
2016/09/16 07:40:45
They need not be an integer multiple of 10.
kwiberg-webrtc
2016/09/16 07:51:57
Are there no restrictions whatsoever, except the u
hlundin-webrtc
2016/09/16 08:25:02
In theory, no other restrictions. In practice I ca
ossu
2016/09/16 11:24:16
Acknowledged.
| |
81 // any frame objects returned by this call. The decoder is free to swap or | |
82 // move the data from the |payload| buffer. |timestamp| is the input | |
83 // timestamp, in samples, corresponding to the start of the payload. | |
84 virtual std::vector<ParseResult> ParsePayload(rtc::Buffer&& payload, | |
85 uint32_t timestamp, | |
86 bool is_primary); | |
87 | |
36 // Decodes |encode_len| bytes from |encoded| and writes the result in | 88 // Decodes |encode_len| bytes from |encoded| and writes the result in |
37 // |decoded|. The maximum bytes allowed to be written into |decoded| is | 89 // |decoded|. The maximum bytes allowed to be written into |decoded| is |
38 // |max_decoded_bytes|. Returns the total number of samples across all | 90 // |max_decoded_bytes|. Returns the total number of samples across all |
39 // channels. If the decoder produced comfort noise, |speech_type| | 91 // channels. If the decoder produced comfort noise, |speech_type| |
40 // is set to kComfortNoise, otherwise it is kSpeech. The desired output | 92 // is set to kComfortNoise, otherwise it is kSpeech. The desired output |
41 // sample rate is provided in |sample_rate_hz|, which must be valid for the | 93 // sample rate is provided in |sample_rate_hz|, which must be valid for the |
42 // codec at hand. | 94 // codec at hand. |
43 int Decode(const uint8_t* encoded, | 95 int Decode(const uint8_t* encoded, |
44 size_t encoded_len, | 96 size_t encoded_len, |
45 int sample_rate_hz, | 97 int sample_rate_hz, |
(...skipping 69 matching lines...) Expand 10 before | Expand all | Expand 10 after Loading... | |
115 int sample_rate_hz, | 167 int sample_rate_hz, |
116 int16_t* decoded, | 168 int16_t* decoded, |
117 SpeechType* speech_type); | 169 SpeechType* speech_type); |
118 | 170 |
119 private: | 171 private: |
120 RTC_DISALLOW_COPY_AND_ASSIGN(AudioDecoder); | 172 RTC_DISALLOW_COPY_AND_ASSIGN(AudioDecoder); |
121 }; | 173 }; |
122 | 174 |
123 } // namespace webrtc | 175 } // namespace webrtc |
124 #endif // WEBRTC_MODULES_AUDIO_CODING_NETEQ_INCLUDE_AUDIO_DECODER_H_ | 176 #endif // WEBRTC_MODULES_AUDIO_CODING_NETEQ_INCLUDE_AUDIO_DECODER_H_ |
OLD | NEW |