Audio Output Example#

Audio Output demonstrates the basic use cases of QAudioOutput.

This example provides a tone generator to supply continuous audio playback. The first button allows pause and resume of the playback, and the second button allows toggling between push and pull modes of operation.

Audio Output Screenshot
"""PySide6 port of the multimedia/audiooutput example from Qt v5.x, originating from PyQt"""

import sys
from math import pi, sin
from struct import pack

from PySide6.QtCore import (QByteArray, QIODevice, Qt, QSysInfo, QTimer,
                            qWarning, Slot)
from PySide6.QtMultimedia import (QAudio, QAudioDevice, QAudioFormat,
                                  QAudioSink, QMediaDevices)
from PySide6.QtWidgets import (QApplication, QComboBox, QHBoxLayout, QLabel,
                               QMainWindow, QPushButton, QSlider,
                               QVBoxLayout, QWidget)


class Generator(QIODevice):

    def __init__(self, format, durationUs, sampleRate, parent):
        super().__init__(parent)

        self.m_pos = 0
        self.m_buffer = QByteArray()

        self.generate_data(format, durationUs, sampleRate)

    def start(self):
        self.open(QIODevice.ReadOnly)

    def stop(self):
        self.m_pos = 0
        self.close()

    def generate_data(self, fmt, durationUs, sampleRate):
        pack_format = ''

        sample_size = fmt.bytesPerSample() * 8
        if sample_size == 8:
            if fmt.sampleFormat() == QAudioFormat.UInt8:
                scaler = lambda x: ((1.0 + x) / 2 * 255)
                pack_format = 'B'
            elif fmt.sampleFormat() == QAudioFormat.Int16:
                scaler = lambda x: x * 127
                pack_format = 'b'
        elif sample_size == 16:
            little_endian = QSysInfo.ByteOrder == QSysInfo.LittleEndian
            if fmt.sampleFormat() == QAudioFormat.UInt8:
                scaler = lambda x: (1.0 + x) / 2 * 65535
                pack_format = '<H' if little_endian else '>H'
            elif fmt.sampleFormat() == QAudioFormat.Int16:
                scaler = lambda x: x * 32767
                pack_format = '<h' if little_endian else '>h'

        assert(pack_format != '')

        channel_bytes = fmt.bytesPerSample()
        sample_bytes = fmt.channelCount() * channel_bytes

        length = (fmt.sampleRate() * fmt.channelCount() * channel_bytes) * durationUs // 100000

        self.m_buffer.clear()
        sample_index = 0
        factor = 2 * pi * sampleRate / fmt.sampleRate()

        while length != 0:
            x = sin((sample_index % fmt.sampleRate()) * factor)
            packed = pack(pack_format, int(scaler(x)))

            for _ in range(fmt.channelCount()):
                self.m_buffer.append(packed)
                length -= channel_bytes

            sample_index += 1

    def readData(self, maxlen):
        data = QByteArray()
        total = 0

        while maxlen > total:
            chunk = min(self.m_buffer.size() - self.m_pos, maxlen - total)
            data.append(self.m_buffer.mid(self.m_pos, chunk))
            self.m_pos = (self.m_pos + chunk) % self.m_buffer.size()
            total += chunk

        return data.data()

    def writeData(self, data):
        return 0

    def bytesAvailable(self):
        return self.m_buffer.size() + super(Generator, self).bytesAvailable()


class AudioTest(QMainWindow):

    PUSH_MODE_LABEL = "Enable push mode"
    PULL_MODE_LABEL = "Enable pull mode"
    SUSPEND_LABEL = "Suspend playback"
    RESUME_LABEL = "Resume playback"

    DURATION_SECONDS = 1
    TONE_SAMPLE_RATE_HZ = 600
    DATA_SAMPLE_RATE_HZ = 44100

    def __init__(self, devices):
        super().__init__()

        self.m_devices = devices
        self.m_device = self.m_devices[0]
        self.m_output = None

        self.initialize_window()
        self.initialize_audio()

    def initialize_window(self):

        central_widget = QWidget()
        layout = QVBoxLayout(central_widget)

        self.m_deviceBox = QComboBox()
        self.m_deviceBox.activated[int].connect(self.device_changed)
        for deviceInfo in self.m_devices:
            self.m_deviceBox.addItem(deviceInfo.description(), deviceInfo)

        layout.addWidget(self.m_deviceBox)

        self.m_modeButton = QPushButton()
        self.m_modeButton.clicked.connect(self.toggle_mode)
        self.m_modeButton.setText(self.PUSH_MODE_LABEL)

        layout.addWidget(self.m_modeButton)

        self.m_suspendResumeButton = QPushButton(
                clicked=self.toggle_suspend_resume)
        self.m_suspendResumeButton.setText(self.SUSPEND_LABEL)

        layout.addWidget(self.m_suspendResumeButton)

        volume_box = QHBoxLayout()
        volume_label = QLabel("Volume:")
        self.m_volumeSlider = QSlider(Qt.Horizontal, minimum=0, maximum=100,
                singleStep=10)
        self.m_volumeSlider.valueChanged.connect(self.volume_changed)

        volume_box.addWidget(volume_label)
        volume_box.addWidget(self.m_volumeSlider)

        layout.addLayout(volume_box)

        self.setCentralWidget(central_widget)

    def initialize_audio(self):
        self.m_pullTimer = QTimer(self)
        self.m_pullTimer.timeout.connect(self.pull_timer_expired)
        self.m_pullMode = True

        self.m_format = QAudioFormat()
        self.m_format.setSampleRate(self.DATA_SAMPLE_RATE_HZ)
        self.m_format.setChannelCount(1)
        self.m_format.setSampleFormat(QAudioFormat.Int16)

        info = self.m_devices[0]
        if not info.isFormatSupported(self.m_format):
            qWarning("Default format not supported - trying to use nearest")
            self.m_format = info.nearestFormat(self.m_format)

        self.m_generator = Generator(self.m_format,
                self.DURATION_SECONDS * 1000000, self.TONE_SAMPLE_RATE_HZ, self)

        self.create_audio_output()

    def create_audio_output(self):
        self.m_audioSink = QAudioSink(self.m_device, self.m_format)
        self.m_audioSink.stateChanged.connect(self.handle_state_changed)

        self.m_generator.start()
        self.m_audioSink.start(self.m_generator)
        self.m_volumeSlider.setValue(self.m_audioSink.volume() * 100)

    @Slot(int)
    def device_changed(self, index):
        self.m_pullTimer.stop()
        self.m_generator.stop()
        self.m_audioSink.stop()
        self.m_device = self.m_deviceBox.itemData(index)

        self.create_audio_output()

    @Slot(int)
    def volume_changed(self, value):
        if self.m_audioSink is not None:
            self.m_audioSink.setVolume(value / 100.0)

    @Slot()
    def notified(self):
        bytes_free = self.m_audioSink.bytesFree()
        elapsed = self.m_audioSink.elapsedUSecs()
        processed = self.m_audioSink.processedUSecs()
        qWarning(f"bytesFree = {bytes_free}, "
                 f"elapsedUSecs = {elapsed}, "
                 f"processedUSecs = {processed}")

    @Slot()
    def pull_timer_expired(self):
        if self.m_audioSink is not None and self.m_audioSink.state() != QAudio.StoppedState:
            bytes_free = self.m_audioSink.bytesFree()
            data = self.m_generator.read(bytes_free)
            if data:
                self.m_output.write(data)

    @Slot()
    def toggle_mode(self):
        self.m_pullTimer.stop()
        self.m_audioSink.stop()

        if self.m_pullMode:
            self.m_modeButton.setText(self.PULL_MODE_LABEL)
            self.m_output = self.m_audioSink.start()
            self.m_pullMode = False
            self.m_pullTimer.start(20)
        else:
            self.m_modeButton.setText(self.PUSH_MODE_LABEL)
            self.m_pullMode = True
            self.m_audioSink.start(self.m_generator)

        self.m_suspendResumeButton.setText(self.SUSPEND_LABEL)

    @Slot()
    def toggle_suspend_resume(self):
        if self.m_audioSink.state() == QAudio.SuspendedState:
            qWarning("status: Suspended, resume()")
            self.m_audioSink.resume()
            self.m_suspendResumeButton.setText(self.SUSPEND_LABEL)
        elif self.m_audioSink.state() == QAudio.ActiveState:
            qWarning("status: Active, suspend()")
            self.m_audioSink.suspend()
            self.m_suspendResumeButton.setText(self.RESUME_LABEL)
        elif self.m_audioSink.state() == QAudio.StoppedState:
            qWarning("status: Stopped, resume()")
            self.m_audioSink.resume()
            self.m_suspendResumeButton.setText(self.SUSPEND_LABEL)
        elif self.m_audioSink.state() == QAudio.IdleState:
            qWarning("status: IdleState")

    state_map = {
        QAudio.ActiveState: "ActiveState",
        QAudio.SuspendedState: "SuspendedState",
        QAudio.StoppedState: "StoppedState",
        QAudio.IdleState: "IdleState"}

    @Slot("QAudio::State")
    def handle_state_changed(self, state):
        state = self.state_map.get(state, 'Unknown')
        qWarning(f"state = {state}")


if __name__ == '__main__':
    app = QApplication(sys.argv)
    app.setApplicationName("Audio Output Test")

    devices = QMediaDevices.audioOutputs()
    if not devices:
        print('No audio outputs found.', file=sys.stderr)
        sys.exit(-1)

    audio = AudioTest(devices)
    audio.show()

    sys.exit(app.exec())