Dont always downsample to 16kHz in the reverse stream in APM

webrtc/modules/audio_processing/audio_processing_impl.cc

Index: webrtc/modules/audio_processing/audio_processing_impl.cc
diff --git a/webrtc/modules/audio_processing/audio_processing_impl.cc b/webrtc/modules/audio_processing/audio_processing_impl.cc
index 66123962c3a87468357b4b006665bd8b6ea6d02d..ed6ad632ca875eba14a3d0eb16c5755f3e50c52b 100644
--- a/webrtc/modules/audio_processing/audio_processing_impl.cc
+++ b/webrtc/modules/audio_processing/audio_processing_impl.cc
@@ -59,6 +59,17 @@
   } while (0)
 
 namespace webrtc {
+
+const int AudioProcessing::kNativeSampleRatesHz[] = {
+    AudioProcessing::kSampleRate8kHz,
+    AudioProcessing::kSampleRate16kHz,
+    AudioProcessing::kSampleRate32kHz,
+    AudioProcessing::kSampleRate48kHz};
+const size_t AudioProcessing::kNumNativeSampleRates =
+    arraysize(AudioProcessing::kNativeSampleRatesHz);
+const int AudioProcessing::kMaxNativeSampleRateHz = AudioProcessing::
+    kNativeSampleRatesHz[AudioProcessing::kNumNativeSampleRates - 1];

[peah-webrtc, 2016/03/10 06:59:24]

Afaics, this means that the tri-band 48 kHz splitting filter can be used on any
platform, right? What about the complexity for that on ARM, or is that addressed
in another CL?

[aluebs-webrtc, 2016/03/10 15:34:32]

That is adressed in this CL that already landed:
https://codereview.webrtc.org/1766103002/
This CL is going to need a rebase and conflict fixing before landing :)

+
 namespace {
 
 static bool LayoutHasKeyboard(AudioProcessing::ChannelLayout layout) {
@@ -74,6 +85,21 @@ static bool LayoutHasKeyboard(AudioProcessing::ChannelLayout layout) {
   assert(false);
   return false;
 }
+
+bool MultiBandSignal(int sample_rate_hz) {
+  return sample_rate_hz == AudioProcessing::kSampleRate32kHz ||
+         sample_rate_hz == AudioProcessing::kSampleRate48kHz;
+}
+
+int ClosestNativeRate(int min_proc_rate) {
+  for (int rate : AudioProcessing::kNativeSampleRatesHz) {
+    if (rate >= min_proc_rate) {
+      return rate;
+    }
+  }
+  return AudioProcessing::kMaxNativeSampleRateHz;
+}
+
 }  // namespace
 
 // Throughout webrtc, it's assumed that success is represented by zero.
@@ -109,16 +135,6 @@ struct AudioProcessingImpl::ApmPrivateSubmodules {
   std::unique_ptr<AgcManagerDirect> agc_manager;
 };
 
-const int AudioProcessing::kNativeSampleRatesHz[] = {
-    AudioProcessing::kSampleRate8kHz,
-    AudioProcessing::kSampleRate16kHz,
-    AudioProcessing::kSampleRate32kHz,
-    AudioProcessing::kSampleRate48kHz};
-const size_t AudioProcessing::kNumNativeSampleRates =
-    arraysize(AudioProcessing::kNativeSampleRatesHz);
-const int AudioProcessing::kMaxNativeSampleRateHz = AudioProcessing::
-    kNativeSampleRatesHz[AudioProcessing::kNumNativeSampleRates - 1];
-
 AudioProcessing* AudioProcessing::Create() {
   Config config;
   return Create(config, nullptr);
@@ -364,32 +380,17 @@ int AudioProcessingImpl::InitializeLocked(const ProcessingConfig& config) {
 
   formats_.api_format = config;
 
-  // We process at the closest native rate >= min(input rate, output rate).
-  const int min_proc_rate =
-      std::min(formats_.api_format.input_stream().sample_rate_hz(),
-               formats_.api_format.output_stream().sample_rate_hz());
-  int fwd_proc_rate;
-  for (size_t i = 0; i < kNumNativeSampleRates; ++i) {
-    fwd_proc_rate = kNativeSampleRatesHz[i];
-    if (fwd_proc_rate >= min_proc_rate) {
-      break;
-    }
-  }
-
-  capture_nonlocked_.fwd_proc_format = StreamConfig(fwd_proc_rate);
+  capture_nonlocked_.fwd_proc_format = StreamConfig(ClosestNativeRate(std::min(
+      formats_.api_format.input_stream().sample_rate_hz(),
+      formats_.api_format.output_stream().sample_rate_hz())));
 
-  // We normally process the reverse stream at 16 kHz. Unless...
-  int rev_proc_rate = kSampleRate16kHz;
+  int rev_proc_rate = ClosestNativeRate(std::min(
+      formats_.api_format.reverse_input_stream().sample_rate_hz(),
+      formats_.api_format.reverse_output_stream().sample_rate_hz()));

[peah-webrtc, 2016/03/10 06:59:24]

This change basically means that the tri-band 48 kHz splitting filter is
activated for the reverse stream, right?

I should know this, but as the CL description says "Dont always downsample to
16kHz in the reverse stream in APM" I'm a bit confused. The code did not always
downsample to 16 lHz before the change either, right? For 32 kHz the 2-band
splitting filter was used, right?

[aluebs-webrtc, 2016/03/10 15:34:32]

That is right. But in that case 32kHz was more of an exception, because the
2-band splitting filter was better than the downsampler.

+  // If the forward sample rate is 8 kHz, the reverse stream is also processed
+  // at this rate.
   if (capture_nonlocked_.fwd_proc_format.sample_rate_hz() == kSampleRate8kHz) {
-    // ...the forward stream is at 8 kHz.
     rev_proc_rate = kSampleRate8kHz;
-  } else {
-    if (formats_.api_format.reverse_input_stream().sample_rate_hz() ==
-        kSampleRate32kHz) {
-      // ...or the input is at 32 kHz, in which case we use the splitting
-      // filter rather than the resampler.
-      rev_proc_rate = kSampleRate32kHz;
-    }
   }
 
   // Always downmix the reverse stream to mono for analysis. This has been
@@ -648,8 +649,7 @@ int AudioProcessingImpl::ProcessStream(AudioFrame* frame) {
 
   capture_.capture_audio->DeinterleaveFrom(frame);
   RETURN_ON_ERR(ProcessStreamLocked());
-  capture_.capture_audio->InterleaveTo(frame,
-                                       output_copy_needed(is_data_processed()));
+  capture_.capture_audio->InterleaveTo(frame, output_copy_needed());
 
 #ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
   if (debug_dump_.debug_file->Open()) {
@@ -689,8 +689,7 @@ int AudioProcessingImpl::ProcessStreamLocked() {
         capture_nonlocked_.fwd_proc_format.num_frames());
   }
 
-  bool data_processed = is_data_processed();
-  if (analysis_needed(data_processed)) {
+  if (fwd_analysis_needed()) {
     ca->SplitIntoFrequencyBands();
   }
 
@@ -729,7 +728,7 @@ int AudioProcessingImpl::ProcessStreamLocked() {
   }
   RETURN_ON_ERR(public_submodules_->gain_control->ProcessCaptureAudio(ca));
 
-  if (synthesis_needed(data_processed)) {
+  if (fwd_synthesis_needed()) {
     ca->MergeFrequencyBands();
   }
 
@@ -904,7 +903,7 @@ int AudioProcessingImpl::AnalyzeReverseStream(AudioFrame* frame) {
 
 int AudioProcessingImpl::ProcessReverseStreamLocked() {
   AudioBuffer* ra = render_.render_audio.get();  // For brevity.
-  if (formats_.rev_proc_format.sample_rate_hz() == kSampleRate32kHz) {
+  if (rev_analysis_needed()) {
     ra->SplitIntoFrequencyBands();
   }
 
@@ -925,8 +924,7 @@ int AudioProcessingImpl::ProcessReverseStreamLocked() {
     RETURN_ON_ERR(public_submodules_->gain_control->ProcessRenderAudio(ra));
   }
 
-  if (formats_.rev_proc_format.sample_rate_hz() == kSampleRate32kHz &&
-      is_rev_processed()) {
+  if (rev_synthesis_needed()) {
     ra->MergeFrequencyBands();
   }
 
@@ -1138,31 +1136,26 @@ bool AudioProcessingImpl::is_data_processed() const {
   return false;
 }
 
-bool AudioProcessingImpl::output_copy_needed(bool is_data_processed) const {
+bool AudioProcessingImpl::output_copy_needed() const {
   // Check if we've upmixed or downmixed the audio.
   return ((formats_.api_format.output_stream().num_channels() !=
            formats_.api_format.input_stream().num_channels()) ||
-          is_data_processed || capture_.transient_suppressor_enabled);
+          is_data_processed() || capture_.transient_suppressor_enabled);
 }
 
-bool AudioProcessingImpl::synthesis_needed(bool is_data_processed) const {
-  return (is_data_processed &&
-          (capture_nonlocked_.fwd_proc_format.sample_rate_hz() ==
-               kSampleRate32kHz ||
-           capture_nonlocked_.fwd_proc_format.sample_rate_hz() ==
-               kSampleRate48kHz));
+bool AudioProcessingImpl::fwd_synthesis_needed() const {
+  return (is_data_processed() &&
+          MultiBandSignal(capture_nonlocked_.fwd_proc_format.sample_rate_hz()));
 }
 
-bool AudioProcessingImpl::analysis_needed(bool is_data_processed) const {
-  if (!is_data_processed &&
+bool AudioProcessingImpl::fwd_analysis_needed() const {
+  if (!is_data_processed() &&
       !public_submodules_->voice_detection->is_enabled() &&
       !capture_.transient_suppressor_enabled) {
     // Only public_submodules_->level_estimator is enabled.
     return false;
-  } else if (capture_nonlocked_.fwd_proc_format.sample_rate_hz() ==
-                 kSampleRate32kHz ||
-             capture_nonlocked_.fwd_proc_format.sample_rate_hz() ==
-                 kSampleRate48kHz) {
+  } else if (MultiBandSignal(
+                 capture_nonlocked_.fwd_proc_format.sample_rate_hz())) {
     // Something besides public_submodules_->level_estimator is enabled, and we
     // have super-wb.
     return true;
@@ -1174,6 +1167,15 @@ bool AudioProcessingImpl::is_rev_processed() const {
   return constants_.intelligibility_enabled;
 }
 
+bool AudioProcessingImpl::rev_synthesis_needed() const {
+  return (is_rev_processed() &&
+          MultiBandSignal(formats_.rev_proc_format.sample_rate_hz()));
+}
+
+bool AudioProcessingImpl::rev_analysis_needed() const {
+  return MultiBandSignal(formats_.rev_proc_format.sample_rate_hz());
+}
+
 bool AudioProcessingImpl::render_check_rev_conversion_needed() const {
   return rev_conversion_needed();
 }