chromium/chrome/services/speech/speech_recognition_recognizer_impl.cc

// Copyright 2020 The Chromium Authors
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

#include "chrome/services/speech/speech_recognition_recognizer_impl.h"

#include <algorithm>
#include <string>
#include <utility>

#include "base/containers/flat_map.h"
#include "base/containers/span.h"
#include "base/files/file_path.h"
#include "base/files/file_util.h"
#include "base/functional/bind.h"
#include "base/metrics/histogram_functions.h"
#include "base/strings/strcat.h"
#include "base/strings/string_util.h"
#include "base/task/bind_post_task.h"
#include "base/task/sequenced_task_runner.h"
#include "base/task/task_runner.h"
#include "base/task/task_traits.h"
#include "base/task/thread_pool.h"
#include "base/time/time.h"
#include "build/build_config.h"
#include "build/chromeos_buildflags.h"
#include "chrome/services/speech/soda/proto/soda_api.pb.h"
#include "chrome/services/speech/soda/soda_client.h"
#include "components/soda/constants.h"
#include "google_apis/google_api_keys.h"
#include "media/base/audio_buffer.h"
#include "media/base/audio_sample_types.h"
#include "media/base/audio_timestamp_helper.h"
#include "media/base/limits.h"
#include "media/base/media_switches.h"
#include "media/mojo/common/media_type_converters.h"
#include "media/mojo/mojom/audio_data.mojom.h"
#include "mojo/public/cpp/bindings/self_owned_receiver.h"

namespace speech {

constexpr char kInvalidAudioDataError[] = …;

// static
const char
    SpeechRecognitionRecognizerImpl::kCaptionBubbleVisibleHistogramName[] = …;

// static
const char
    SpeechRecognitionRecognizerImpl::kCaptionBubbleHiddenHistogramName[] = …;

constexpr char kLiveCaptionLanguageCountHistogramName[] = …;

namespace {

// Callback executed by the SODA library on a speech recognition event. The
// callback handle is a void pointer to the SpeechRecognitionRecognizerImpl that
// owns the SODA instance. SpeechRecognitionRecognizerImpl owns the SodaClient
// which owns the instance of SODA and their sequential destruction order
// ensures that this callback will never be called with an invalid callback
// handle to the SpeechRecognitionRecognizerImpl.
void OnSodaResponse(const char* serialized_proto,
                    int length,
                    void* callback_handle) { … }

speech::soda::chrome::ExtendedSodaConfigMsg::RecognitionMode
GetSodaSpeechRecognitionMode(
    media::mojom::SpeechRecognitionMode recognition_mode) { … }

}  // namespace

SpeechRecognitionRecognizerImpl::~SpeechRecognitionRecognizerImpl() { … }

void SpeechRecognitionRecognizerImpl::OnLanguagePackInstalled(
    base::flat_map<std::string, base::FilePath> config_paths) { … }

void SpeechRecognitionRecognizerImpl::Create(
    mojo::PendingReceiver<media::mojom::SpeechRecognitionRecognizer> receiver,
    mojo::PendingRemote<media::mojom::SpeechRecognitionRecognizerClient> remote,
    media::mojom::SpeechRecognitionOptionsPtr options,
    const base::FilePath& binary_path,
    const base::flat_map<std::string, base::FilePath>& config_paths,
    const std::string& primary_language_name,
    const bool mask_offensive_words,
    base::WeakPtr<SpeechRecognitionServiceImpl> speech_recognition_service) { … }

bool SpeechRecognitionRecognizerImpl::IsMultichannelSupported() { … }

void SpeechRecognitionRecognizerImpl::OnRecognitionEvent(
    media::SpeechRecognitionResult event) { … }

void SpeechRecognitionRecognizerImpl::
    OnSpeechRecognitionRecognitionEventCallback(bool success) { … }

void SpeechRecognitionRecognizerImpl::OnLanguageIdentificationEvent(
    const std::string& language,
    const media::mojom::ConfidenceLevel confidence_level,
    const media::mojom::AsrSwitchResult asr_switch_result) { … }

void SpeechRecognitionRecognizerImpl::OnRecognitionStoppedCallback() { … }

SpeechRecognitionRecognizerImpl::SpeechRecognitionRecognizerImpl(
    mojo::PendingRemote<media::mojom::SpeechRecognitionRecognizerClient> remote,
    media::mojom::SpeechRecognitionOptionsPtr options,
    const base::FilePath& binary_path,
    const base::flat_map<std::string, base::FilePath>& config_paths,
    const std::string& primary_language_name,
    const bool mask_offensive_words,
    base::WeakPtr<SpeechRecognitionServiceImpl> speech_recognition_service)
    : … { … }

void SpeechRecognitionRecognizerImpl::OnClientHostDisconnected() { … }

void SpeechRecognitionRecognizerImpl::SendAudioToSpeechRecognitionService(
    media::mojom::AudioDataS16Ptr buffer) { … }

void SpeechRecognitionRecognizerImpl::OnSpeechRecognitionError() { … }

void SpeechRecognitionRecognizerImpl::MarkDone() { … }

void SpeechRecognitionRecognizerImpl::AddAudio(
    media::mojom::AudioDataS16Ptr buffer) { … }
void SpeechRecognitionRecognizerImpl::OnAudioCaptureEnd() { … }
void SpeechRecognitionRecognizerImpl::OnAudioCaptureError() { … }

void SpeechRecognitionRecognizerImpl::
    SendAudioToSpeechRecognitionServiceInternal(
        media::mojom::AudioDataS16Ptr buffer) { … }

void SpeechRecognitionRecognizerImpl::OnLanguageChanged(
    const std::string& language) { … }

void SpeechRecognitionRecognizerImpl::OnMaskOffensiveWordsChanged(
    bool mask_offensive_words) { … }

void SpeechRecognitionRecognizerImpl::ResetSodaWithNewLanguage(
    std::string language_name,
    std::pair<base::FilePath, bool> config_and_exists) { … }

void SpeechRecognitionRecognizerImpl::RecordDuration() { … }

void SpeechRecognitionRecognizerImpl::ResetSoda() { … }

}  // namespace speech