Signal Generator tool

This directory contains a developer tool that generates and outputs audio signals, via the Audio, AudioCore, AudioRenderer, VolumeControl and GainControl FIDL protocols.

USAGE

signal_generator
  [--chans=<NUM_CHANS>]
  [--int16 | --int24]
  [--rate=<FRAME_RATE>]
  [--sine[=<FREQ>] | --square[=<FREQ>] | --pulse[=<FREQ>] | --saw[=<FREQ>] | --tri[=<FREQ>]
    | --noise | --pink]
  [--dur=<DURATION_SEC>]
  [--amp[=<AMPL>]]
  [--wav[=<FILEPATH>]]
  [--usage=<RENDER_USAGE>]
  [--usage-vol[=<USAGE_VOLUME>]]
  [--usage-gain[=<USAGE_GAIN_DB>]]
  [--flexible-clock | --monotonic-clock | --custom-clock | --rate-adjust[=<PPM>]]
  [--ref]
  [--media[=<PTS>]]
  [--pts]
  [--threshold=<SECS>]
  [--frames=<FRAMES>]
  [--num-bufs=<BUFFERS>]
  [--buffer=<FRAMES>]
  [--online]
  [--gain[=<STREAM_GAIN_DB>]]
  [--mute[=<0|1>]]
  [--ramp]
  [--end-gain=<RAMP_END_DB>]
  [--ramp-dur=<RAMP_DURATION_MSEC>]
  [--ultrasound]
  [--v]
  [--help | --?]

These optional parameters are interpreted as follows:

  By default, stream format is 2-channel, float32 samples at 48000 Hz frame rate
--chans=<NUM_CHANS>      Specify number of channels
--int16                  Use 16-bit integer samples
--int24                  Use 24-in-32-bit integer samples (left-justified 'padded-24')
--rate=<FRAME_RATE>      Set frame rate in Hz

  By default, signal is a sine wave. If no frequency is provided, 440.0 Hz is used
--sine[=<FREQ>]          Play sine wave at given frequency (Hz)
--pulse[=<FREQ>]         Play variable-duty-cycle pulse wave at given frequency
--square[=<FREQ>]        Play square wave at given frequency
                         (equivalent to '--pulse' with '--duty=50.0')
--saw[=<FREQ>]           Play rising sawtooth wave at given frequency
--tri[=<FREQ>]           Play rising-then-falling triangle wave at given frequency
--noise                  Play pseudo-random 'white' noise
--pink                   Play pseudo-random 'pink' (1/f) noise

  By default, play signal for 2.0 seconds, at amplitude 0.25
--dur=<DURATION_SECS>    Set playback length, in seconds
--amp[=<AMPL>]           Set amplitude (silence=0.0, full-scale=1.0, 1.0 if no value provided)
--duty[=<PERCENT>]       Set duty cycle, in percent. Only for pulse waves.
                         (50.0 if only '--duty')

--wav[=<FILEPATH>]       Save to .wav file (default '/tmp/signal_generator.wav')

  Subsequent settings (e.g. gain, timestamps) do not affect .wav file contents

  By default, use a MEDIA stream and do not change the volume or gain for this RENDER_USAGE
--usage=<RENDER_USAGE>   Set stream render usage. RENDER_USAGE must be one of:
                         ACCESSIBILITY, BACKGROUND, MEDIA, INTERRUPTION, SYSTEM_AGENT, COMMUNICATION
--usage-vol[=<VOLUME>]   Set render usage volume (min 0.0, max 1.0, 1.0 if flag with no value)
--usage-gain[=<DB>]      Set render usage gain, in dB (min -160.0, max 0.0, default 0.0)
  Changes to these system-wide volume/gain settings persist after the utility runs

  Use the default reference clock unless specified otherwise
--flexible-clock         Request and use the 'flexible' reference clock provided by the Audio service
--monotonic-clock        Clone CLOCK_MONOTONIC and use it as this stream's reference clock
--custom-clock           Create and use a custom clock as this stream's reference clock
--rate-adjust[=<PPM>]    Run faster/slower than local system clock, in parts-per-million
                         (-1000 min, +1000 max, use -75 if unspecified). Implies '--custom-clock'

  By default, submit data in non-timestamped buffers of 480 frames and 1 VMO,
  without specifying a precise reference time or PTS for the start of playback
--ref                    Specify a reference time in the Play() method
--media[=<PTS>]          Use a specifie PTS value for playback start
--pts                    Apply presentation timestamps (units: frames)
--threshold[=<SECS>]     Set PTS discontinuity threshold, in seconds (default 0.000125)
--frames=<FRAMES>        Set packet size, in frames
--num-bufs=<BUFFERS>     Set the number of payload buffers
--buffer=<FRAMES>        Set size of each payload buffer, in frames
                         Payload buffer space must exceed renderer MinLeadTime or signal duration

  By default, submit packets upon previous packet completions
--online                Emit packets at precisely calculated times, ignoring previous completions.
                        This simulates playback from an external source, such as a network.
                        (This doubles the payload buffer space requirement mentioned above.)

  By default, do not set AudioRenderer gain/mute (unity 0.0 dB, unmuted, no ramping)
--gain[=<GAIN_DB>]       Set stream gain, in dB (min -160.0, max 24.0, default 0.0)
--mute[=<0|1>]           Set stream mute (0=Unmute or 1=Mute; Mute if only '--mute' is provided)
--ramp                   Smoothly ramp gain from initial value to target -75.0 dB by end-of-signal
                         If '--gain' is not provided, ramping starts at unity stream gain (0.0 dB)
--end-gain=<TARGET_DB>   Set a different ramp target gain (dB). Implies '--ramp'
--ramp-dur=<MSECS>       Set a specific ramp duration in milliseconds. Implies '--ramp'

--ultrasound             Play signal using an ultrasound renderer

--v                      Display per-packet information
--help, --?              Show this message

IMPORTANT NOTE

Developers can use this tool to change systemwide render-usage volume and gain. These changes persist beyond this tool's invocation.