VoiceRecognitionLayer Class Reference

A security layer simulating voice recognition verification. More...

#include <voice_recognition_layer.h>

Inheritance diagram for VoiceRecognitionLayer:

Collaboration diagram for VoiceRecognitionLayer:

Public Member Functions
void	Initialize () override
	Initializes the layer for display. Resets the layer state, ensures the layer is fully opaque, and starts audio recording after a short delay.
void	Reset () override
	Resets the layer to its initial state. Stops recording and timers, resets progress, clears audio buffers and visualizer data, resets UI element styles, and ensures the layer is fully opaque.
	VoiceRecognitionLayer (QWidget *parent=nullptr)
	Constructs a VoiceRecognitionLayer. Initializes the UI elements (title, audio visualizer, instructions, progress bar), sets up the layout, creates timers for progress, recognition duration, and audio processing, and initializes audio-related variables.
	~VoiceRecognitionLayer () override
	Destructor. Stops audio recording and processing, stops and deletes timers.
Public Member Functions inherited from SecurityLayer
	SecurityLayer (QWidget *parent=nullptr)
	Constructs a SecurityLayer.
	~SecurityLayer () override=default
	Virtual destructor. Ensures proper cleanup for derived classes.

Private Slots
void	FinishRecognition ()
	Slot called when the recognition process completes (either by progress reaching 100% or recognition_timer_ timeout). Stops recording and timers, updates UI elements to green to indicate success, renders a final green waveform, and initiates the fade-out animation after a short delay.
void	ProcessAudioInput ()
	Slot called periodically by audio_process_timer_ to read and process available audio data. Reads data from the audio input device, calculates the current audio level, determines if the user is speaking (using a threshold and hysteresis), updates the visualizer data, and appends audio data to the buffer if speech is detected.
void	UpdateProgress ()
	Slot called periodically by progress_timer_ to update the recognition progress bar. Increments the progress bar value only if speech is currently detected (is_speaking_ is true). Calls FinishRecognition() when the progress reaches 100%.

Private Member Functions
bool	IsSpeaking (float audio_level) const
	Determines if the given audio level exceeds the noise threshold.
void	StartRecording ()
	Starts capturing audio from the default input device. Configures the audio format, initializes QAudioInput, starts the input device, and starts the associated timers (progress, recognition duration, audio processing).
void	StopRecording ()
	Stops capturing audio and processing. Stops the QAudioInput, deletes the audio input object, stops the audio processing timer, and sets the is_recording_ flag to false.
void	UpdateAudioVisualizer (const QByteArray &data)
	Updates the audio_visualizer_label_ with a waveform representation of the audio data. Shifts existing visualizer data, adds new amplitude data, and redraws the waveform using QPainter. The color of the waveform changes based on whether speech is detected. Also draws a noise threshold indicator and a status text.

Private Attributes
QByteArray	audio_buffer_
	Buffer storing raw audio data captured while the user is speaking.
QIODevice *	audio_device_
	I/O device providing access to the raw audio stream from audio_input_.
QAudioInput *	audio_input_
	Object managing audio input capture.
QTimer *	audio_process_timer_
	Timer controlling how often audio input is processed.
QLabel *	audio_visualizer_label_
	Label used as a canvas to draw the audio waveform visualization.
float	current_audio_level_
	The calculated audio level from the most recently processed audio chunk.
bool	is_recording_
	Flag indicating if audio is currently being recorded.
bool	is_speaking_
	Flag indicating if speech is currently detected (based on threshold and hysteresis).
float	noise_threshold_
	Threshold level above which audio input is considered potential speech.
QTimer *	progress_timer_
	Timer controlling the progress bar update based on speech detection.
QProgressBar *	recognition_progress_
	Progress bar indicating the voice recognition progress.
QTimer *	recognition_timer_
	Timer defining the maximum duration for the recognition attempt.
int	silence_counter_
	Counter tracking the duration of silence in milliseconds.
TranslationManager *	translator_
	The translation manager for handling UI translations.
QVector< float >	visualizer_data_
	Vector storing normalized amplitude values for rendering the waveform visualizer.

Additional Inherited Members
Signals inherited from SecurityLayer
void	layerCompleted ()
	Emitted when the user successfully completes the security challenge presented by this layer.
void	layerFailed ()
	Emitted when the user fails the security challenge (e.g., incorrect code, timeout, collision). Note: Not all layers currently implement failure conditions that emit this signal.

Detailed Description

A security layer simulating voice recognition verification.

This layer captures audio input from the default microphone and displays a real-time audio waveform visualization. It requires the user to speak continuously for a set duration (simulated by a progress bar filling up while speech is detected above a noise threshold). Upon successful completion (progress reaches 100%), the UI elements turn green, and the layer fades out, emitting the layerCompleted() signal.

Constructor & Destructor Documentation

VoiceRecognitionLayer()

VoiceRecognitionLayer::VoiceRecognitionLayer ( QWidget * parent = nullptr )

explicit

Constructs a VoiceRecognitionLayer. Initializes the UI elements (title, audio visualizer, instructions, progress bar), sets up the layout, creates timers for progress, recognition duration, and audio processing, and initializes audio-related variables.

Parameters

parent

Optional parent widget.

Here is the call graph for this function:

~VoiceRecognitionLayer()

VoiceRecognitionLayer::~VoiceRecognitionLayer ( )

override

Destructor. Stops audio recording and processing, stops and deletes timers.

Here is the call graph for this function:

Member Function Documentation

FinishRecognition

void VoiceRecognitionLayer::FinishRecognition ( )

privateslot

Slot called when the recognition process completes (either by progress reaching 100% or recognition_timer_ timeout). Stops recording and timers, updates UI elements to green to indicate success, renders a final green waveform, and initiates the fade-out animation after a short delay.

Here is the call graph for this function:

Here is the caller graph for this function:

Initialize()

void VoiceRecognitionLayer::Initialize ( )

overridevirtual

Initializes the layer for display. Resets the layer state, ensures the layer is fully opaque, and starts audio recording after a short delay.

Implements SecurityLayer.

Here is the call graph for this function:

IsSpeaking()

bool VoiceRecognitionLayer::IsSpeaking ( float audio_level ) const

private

Determines if the given audio level exceeds the noise threshold.

Parameters

audio_level

The calculated audio level (typically normalized RMS or average absolute value).

Returns

True if the level is above the threshold, false otherwise.

Note

Actual speech detection logic with hysteresis is implemented in ProcessAudioInput().

ProcessAudioInput

void VoiceRecognitionLayer::ProcessAudioInput ( )

privateslot

Slot called periodically by audio_process_timer_ to read and process available audio data. Reads data from the audio input device, calculates the current audio level, determines if the user is speaking (using a threshold and hysteresis), updates the visualizer data, and appends audio data to the buffer if speech is detected.

Here is the call graph for this function:

Here is the caller graph for this function:

Reset()

void VoiceRecognitionLayer::Reset ( )

overridevirtual

Resets the layer to its initial state. Stops recording and timers, resets progress, clears audio buffers and visualizer data, resets UI element styles, and ensures the layer is fully opaque.

Implements SecurityLayer.

Here is the call graph for this function:

Here is the caller graph for this function:

StartRecording()

void VoiceRecognitionLayer::StartRecording ( )

private

Starts capturing audio from the default input device. Configures the audio format, initializes QAudioInput, starts the input device, and starts the associated timers (progress, recognition duration, audio processing).

Here is the caller graph for this function:

StopRecording()

void VoiceRecognitionLayer::StopRecording ( )

private

Stops capturing audio and processing. Stops the QAudioInput, deletes the audio input object, stops the audio processing timer, and sets the is_recording_ flag to false.

Here is the caller graph for this function:

UpdateAudioVisualizer()

void VoiceRecognitionLayer::UpdateAudioVisualizer ( const QByteArray & data )

private

Updates the audio_visualizer_label_ with a waveform representation of the audio data. Shifts existing visualizer data, adds new amplitude data, and redraws the waveform using QPainter. The color of the waveform changes based on whether speech is detected. Also draws a noise threshold indicator and a status text.

Parameters

data	The raw audio data chunk to visualize (used to calculate the new amplitude).

Here is the caller graph for this function:

UpdateProgress

void VoiceRecognitionLayer::UpdateProgress ( )

privateslot

Slot called periodically by progress_timer_ to update the recognition progress bar. Increments the progress bar value only if speech is currently detected (is_speaking_ is true). Calls FinishRecognition() when the progress reaches 100%.

Here is the call graph for this function:

Here is the caller graph for this function:

Member Data Documentation

audio_buffer_

QByteArray VoiceRecognitionLayer::audio_buffer_

private

Buffer storing raw audio data captured while the user is speaking.

audio_device_

QIODevice* VoiceRecognitionLayer::audio_device_

private

I/O device providing access to the raw audio stream from audio_input_.

audio_input_

QAudioInput* VoiceRecognitionLayer::audio_input_

private

Object managing audio input capture.

audio_process_timer_

QTimer* VoiceRecognitionLayer::audio_process_timer_

private

Timer controlling how often audio input is processed.

audio_visualizer_label_

QLabel* VoiceRecognitionLayer::audio_visualizer_label_

private

Label used as a canvas to draw the audio waveform visualization.

current_audio_level_

float VoiceRecognitionLayer::current_audio_level_

private

The calculated audio level from the most recently processed audio chunk.

is_recording_

bool VoiceRecognitionLayer::is_recording_

private

Flag indicating if audio is currently being recorded.

is_speaking_

bool VoiceRecognitionLayer::is_speaking_

private

Flag indicating if speech is currently detected (based on threshold and hysteresis).

noise_threshold_

float VoiceRecognitionLayer::noise_threshold_

private

Threshold level above which audio input is considered potential speech.

progress_timer_

QTimer* VoiceRecognitionLayer::progress_timer_

private

Timer controlling the progress bar update based on speech detection.

recognition_progress_

QProgressBar* VoiceRecognitionLayer::recognition_progress_

private

Progress bar indicating the voice recognition progress.

recognition_timer_

QTimer* VoiceRecognitionLayer::recognition_timer_

private

Timer defining the maximum duration for the recognition attempt.

silence_counter_

int VoiceRecognitionLayer::silence_counter_

private

Counter tracking the duration of silence in milliseconds.

translator_

TranslationManager* VoiceRecognitionLayer::translator_

private

The translation manager for handling UI translations.

visualizer_data_

QVector<float> VoiceRecognitionLayer::visualizer_data_

private

Vector storing normalized amplitude values for rendering the waveform visualizer.

---

The documentation for this class was generated from the following files:

• src/ui/settings/tabs/classified/layers/voice_recognition/voice_recognition_layer.h
• src/ui/settings/tabs/classified/layers/voice_recognition/voice_recognition_layer.cpp