Make AudioDebugFileWriter create its own file worker thread.

content/browser/speech/speech_recognizer_impl.cc

 // Copyright (c) 2013 The Chromium Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #include "content/browser/speech/speech_recognizer_impl.h"
 
 #include <stdint.h>
 
 #include <algorithm>
 
 #include "base/bind.h"
 #include "base/macros.h"
+#include "base/task_scheduler/post_task.h"
 #include "base/time/time.h"
 #include "build/build_config.h"
 #include "content/browser/browser_main_loop.h"
 #include "content/browser/media/media_internals.h"
 #include "content/browser/speech/audio_buffer.h"
 #include "content/public/browser/speech_recognition_event_listener.h"
 #include "media/audio/audio_manager.h"
 #include "media/audio/audio_system.h"
 #include "media/base/audio_converter.h"
 
@@ skipping 605 matching lines @@
     input_parameters.set_frames_per_buffer(frames_per_buffer);
     DVLOG(1) << "SRI::input_parameters: "
              << input_parameters.AsHumanReadableString();
   }
 
   // Create an audio converter which converts data between native input format
   // and WebSpeech specific output format.
   audio_converter_.reset(
       new OnDataConverter(input_parameters, output_parameters));
 
   audio_controller_ = AudioInputController::Create(

[fdoray, 2017/05/17 14:58:52]

The TaskRunner ends up being used here:
https://cs.chromium.org/chromium/src/media/audio/audio_debug_file_writer.h. We
shouldn't pass TaskRunners through several components. Instead, the component
that uses a TaskRunner should be the one that creates it. 

I think it would be worth it to remove the TaskRunner argument from all function
calls between here and the constructor of AudioDebugFileWriter. Then, we could
add a link to this CL in
https://chromium.googlesource.com/chromium/src/+/781747b0aa041e426b6446bd49c1...
to demonstrate best practices. (You will get points for all the files that you
have touched + 5 extra points for adding a link to the CL in the documentation).

       GetAudioManager(), this, this, nullptr, input_parameters, device_id_,
       /*agc_is_enabled*/ false,
-      BrowserThread::GetTaskRunnerForThread(BrowserThread::FILE));
+      base::CreateSingleThreadTaskRunnerWithTraits(
+          {base::MayBlock(), base::TaskPriority::BACKGROUND}));
 
   if (!audio_controller_.get()) {
     return Abort(
         SpeechRecognitionError(SPEECH_RECOGNITION_ERROR_AUDIO_CAPTURE));
   }
 
   audio_log_->OnCreated(0, input_parameters, device_id_);
 
   // The endpointer needs to estimate the environment/background noise before
   // starting to treat the audio as user input. We wait in the state
@@ skipping 254 matching lines @@
       engine_error(SPEECH_RECOGNITION_ERROR_NONE) {
 }
 
 SpeechRecognizerImpl::FSMEventArgs::FSMEventArgs(const FSMEventArgs& other) =
     default;
 
 SpeechRecognizerImpl::FSMEventArgs::~FSMEventArgs() {
 }
 
 }  // namespace content