Audio Server

Server

class Server(sr=44100, nchnls=2, buffersize=256, duplex=1, audio='portaudio', jackname='pyo', ichnls=None, winhost='directsound', midi='portmidi', verbosity=7)

Main processing audio loop callback handler.

The Server object handles all communications with Portaudio and Portmidi. It keeps track of all audio streams created as well as connections between them.

An instance of the Server must be booted before defining any signal processing chain.

Args

sr: int, optional

Sampling rate used by Portaudio and the Server to compute samples. Defaults to 44100.

nchnls: int, optional

Number of output channels. The number of input channels will be the same if ichnls argument is not defined. Defaults to 2.

buffersize: int, optional

Number of samples that Portaudio will request from the callback loop. Defaults to 256.

This value has an impact on CPU use (a small buffer size is harder to compute) and on the latency of the system.

Latency is buffer size / sampling rate in seconds.

duplex: int {0, 1}, optional

Input - output mode. 0 is output only and 1 is both ways. Defaults to 1.

audio: string {‘portaudio’, ‘pa’, ‘jack’, ‘coreaudio’, ‘offline’, ‘offline_nb’, ‘embedded’, ‘manual’}, optional

Audio backend to use. ‘pa’ is equivalent to ‘portaudio’. Default is ‘portaudio’.

‘offline’ save the audio output in a soundfile as fast as possible in blocking mode,

ie. the main program doesn’t respond until the end of the computation.

‘offline_nb’ save the audio output in a soundfile as fast as possible in non-blocking mode,

ie. the computation is executed in a separated thread, allowing the program to respond while the computation goes on.

It is the responsibility of the user to make sure that the program doesn’t exit before the computation is done.

‘embedded’ should be used when pyo is embedded inside an host environment via its C api.

If ‘jack’ is selected but jackd is not already started when the program is executed, pyo will ask jack to start in the background. Note that pyo never ask jack to close. It is the user’s responsability to manage the audio configuration of its system.

If ‘manual’ is selected, the server waits for a call to its process method to compute a single buffer size of audio samples. Successive calls to process can simulate a real computation (probably only useful in the context of internal testing).

User can set an environment variable named PYO_SERVER_AUDIO to set this value globally.

jackname: string, optional

Name of jack client. Defaults to ‘pyo’

ichnls: int, optional

Number of input channels if different of output channels. If None (default), ichnls = nchnls.

winhost: string, optional

Under Windows, pyo’s Server will try to use the default devices of the given host. This behaviour can be changed with the SetXXXDevice methods. Defaults to “directsound”.

User can set an environment variable named PYO_SERVER_WINHOST to set this value globally.

midi: string {‘portmidi’, ‘pm’, ‘jack’}, optional

Midi backend to use. ‘pm’ is equivalent to ‘portmidi’. Default is ‘portmidi’.

If ‘jack’ is selected but jackd is not already started when the program is executed, pyo will ask jack to start in the background. Note that pyo never ask jack to close. It is the user’s responsability to manage the audio/midi configuration of its system.

User can set an environment variable named PYO_SERVER_MIDI to set this value globally.

verbosity: int, optional

Set the server’s verbosity.

Note

The following methods must be called before booting the server

• setInOutDevice(x): Set both input and output devices. See pa_list_devices().

• setInputDevice(x): Set the audio input device number. See pa_list_devices().

• setOutputDevice(x): Set the audio output device number. See pa_list_devices().

• setInputOffset(x): Set the first physical input channel.

• setOutputOffset(x): Set the first physical output channel.

• setInOutOffset(x): Set the first physical input and output channels.

• setMidiInputDevice(x): Set the MIDI input device number. See pm_list_devices().

• setMidiOutputDevice(x): Set the MIDI output device number. See pm_list_devices().

• setSamplingRate(x): Set the sampling rate used by the server.

• setBufferSize(x): Set the buffer size used by the server.

• setNchnls(x): Set the number of output (and input if ichnls = None) channels used by the server.

• setIchnls(x): Set the number of input channels (if different of output channels) used by the server.

• setDuplex(x): Set the duplex mode used by the server.

• setVerbosity(x): Set the server’s verbosity.

• reinit(sr, nchnls, buffersize, duplex, audio, jackname): Reinit the server’s settings.

• deactivateMidi(): Deactivate Midi callback.

• setIsJackTransportSlave(x): Set if pyo’s server is slave to jack transport or not.

• allowMicrosoftMidiDevices(): Allows the Microsoft Midi Mapper or GS Wavetable Synth devices.

>>> # For an 8 channels server in duplex mode with
>>> # a sampling rate of 48000 Hz and buffer size of 512
>>> s = Server(sr=48000, nchnls=8, buffersize=512, duplex=1).boot()
>>> s.start()

Public Data Attributes:

amp	float.
startoffset	float.
verbosity	int.
globalseed	int.

Public Methods:

__init__([sr, nchnls, buffersize, duplex, ...])
__del__()
reinit([sr, nchnls, buffersize, duplex, ...])	Reinit the server'settings.
setCallback(callback)	Register a custom process callback.
gui([locals, meter, timer, exit, title])	Show the server's user interface.
closeGui()	Programmatically close the server's GUI.
setTimeCallable(func)	Set a function callback that will receive the current time as argument.
setMeterCallable(func)	Set a function callback that will receive the current rms values as argument.
setMeter(meter)	Registers a meter object to the server.
setInOutDevice(x)	Set both input and output audio devices.
setInputDevice(x)	Set the audio input device number.
setOutputDevice(x)	Set the audio output device number.
setInputOffset(x)	Set the first physical input channel.
setOutputOffset(x)	Set the first physical output channel.
setInOutOffset(x)	Set the first physical input and output channels.
setMidiInputDevice(x)	Set the Midi input device number.
setMidiOutputDevice(x)	Set the Midi output device number.
allowMicrosoftMidiDevices()	Allows the Microsoft Midi Mapper or GS Wavetable Synth device.
setSamplingRate(x)	Set the sampling rate used by the server.
setBufferSize(x)	Set the buffer size used by the server.
setNchnls(x)	Set the number of output (and input if ichnls = None) channels used by the server.
setIchnls(x)	Set the number of input channels (if different of output channels) used by the server.
setDuplex(x)	Set the duplex mode used by the server.
setVerbosity(x)	Set the server's verbosity.
setGlobalDur(x)	Set the global object duration (time to wait before stopping the object).
setGlobalDel(x)	Set the global object delay time (time to wait before activating the object).
setJackAuto([xin, xout])	Tells the server to auto-connect (or not) Jack ports to System ports.
setJackAutoConnectInputPorts(ports)	Tells the server to auto-connect Jack input ports to pre-defined Jack ports.
setJackAutoConnectOutputPorts(ports)	Tells the server to auto-connect Jack output ports to pre-defined Jack ports.
setJackInputPortNames(name)	Change the short name of pyo's input ports for the jack server.
setJackOutputPortNames(name)	Change the short name of pyo's output ports for the jack server.
setJackAutoConnectMidiInputPort(ports)	Tells the server to auto-connect Jack midi input port to pre-defined Jack ports.
setJackAutoConnectMidiOutputPort(ports)	Tells the server to auto-connect Jack midi output port to pre-defined Jack ports.
setJackMidiInputPortName(name)	Change the short name of pyo's midi input port for the jack server.
setJackMidiOutputPortName(name)	Change the short name of pyo's midi output port for the jack server.
setIsJackTransportSlave(x)	Set if pyo's server is slave to jack transport or not.
setGlobalSeed(x)	Set the server's global seed used by random objects.
setStartOffset(x)	Set the server's starting time offset.
setAmp(x)	Set the overall amplitude.
beginResamplingBlock(x)	Starts a resampling block.
endResamplingBlock()	Ends a resampling block.
shutdown()	Shut down and clear the server.
boot([newBuffer])	Boot the server.
start()	Start the audio callback loop and begin processing.
stop()	Stop the audio callback loop.
recordOptions([dur, filename, fileformat, ...])	Sets options for soundfile created by offline rendering or global recording.
recstart([filename])	Begins a default recording of the sound that is sent to the soundcard.
recstop()	Stop the previously started recording.
noteout(pitch, velocity[, channel, timestamp])	Send a MIDI note message to the selected midi output device.
afterout(pitch, velocity[, channel, timestamp])	Send an aftertouch message to the selected midi output device.
ctlout(ctlnum, value[, channel, timestamp])	Send a control change message to the selected midi output device.
programout(value[, channel, timestamp])	Send a program change message to the selected midi output device.
pressout(value[, channel, timestamp])	Send a channel pressure message to the selected midi output device.
bendout(value[, channel, timestamp])	Send a pitch bend message to the selected midi output device.
sysexout(msg[, timestamp])	Send a system exclusive message to the selected midi output device.
makenote(pitch, velocity, duration[, channel])	Send MIDI noteon/noteoff messages to the selected midi output device.
addMidiEvent(status[, data1, data2])	Add a MIDI event in the server processing loop.
getStreams()	Returns the list of Stream objects currently in the Server memory.
getSamplingRate()	Return the current sampling rate.
getNchnls()	Return the current number of channels.
getBufferSize()	Return the current buffer size.
getGlobalDur()	Return the current global duration.
getGlobalDel()	Return the current global delay time.
getGlobalSeed()	Return the current global seed.
getIsStarted()	Returns 1 if the server is started, otherwise returns 0.
getIsBooted()	Returns 1 if the server is booted, otherwise returns 0.
deactivateMidi()	Deactivate Midi callback.
getMidiActive()	Returns 1 if Midi callback is active, otherwise returns 0.
getNumberOfStreams()	Returns the number of streams currently in the Server memory.
setServer()	Sets this server as the one to use for new objects when using the embedded device.
getInputAddr()	Return the address of the input buffer.
getOutputAddr()	Return the address of the output buffer.
getServerID()	Return the server ID.
getServerAddr()	Return the address of the server.
getEmbedICallbackAddr()	Return the address of the interleaved embedded callback function.
getCurrentTime()	Return the current time as a formatted string.
getCurrentTimeInSamples()	Return the current time in number of elapsed samples since the server was started.
getCurrentAmp()	Return the current amplitudes as a tuple of nchnls length.
setAutoStartChildren(state)	Giving True to this method tells pyo that a call to the play, out or stop method of audio objects should propagate to the other audio objects given as arguments.
process()	Tell the server to compute a single buffer size of audio samples.

---

reinit(sr=44100, nchnls=2, buffersize=256, duplex=1, audio='portaudio', jackname='pyo', ichnls=None, winhost='directsound', midi='portmidi')

Reinit the server’settings. Useful to alternate between real-time and offline server.

Args

Same as in the __init__ method.

setCallback(callback)

Register a custom process callback.

The function given as argument will be called every computation block, just before the computation of the audio object tree. Inside the callback, one can process the data of a table with numpy calls for example.

gui(locals=None, meter=True, timer=True, exit=True, title=None)

Show the server’s user interface.

Args

locals: locals namespace {locals(), None}, optional

If locals() is given, the interface will show an interpreter extension, giving a way to interact with the running script. Defaults to None.

meter: boolean, optinal

If True, the interface will show a vumeter of the global output signal. Defaults to True.

timer: boolean, optional

If True, the interface will show a clock of the current time. Defaults to True.

exit: boolean, optional

If True, the python interpreter will exit when the ‘Quit’ button is pressed, Otherwise, the GUI will be closed leaving the interpreter alive. Defaults to True.

title: str, optional

Alternate title for the server window. If None (default), generic title, “Pyo Server” is used.

closeGui()

Programmatically close the server’s GUI.

setTimeCallable(func)

Set a function callback that will receive the current time as argument.

The function will receive four integers in this format:

hours, minutes, seconds, milliseconds

Args

func: python callable

Python function or method to call with current time as argument.

setMeterCallable(func)

Set a function callback that will receive the current rms values as argument.

The function will receive a list containing the rms value for each audio channel.

Args

func: python callable

Python function or method to call with current rms values as argument.

setMeter(meter)

Registers a meter object to the server.

The object must have a method named setRms. This method will be called with the rms values of each audio channel as argument.

Args

meter: python object

Python object with a setRms method.

setInOutDevice(x)

Set both input and output audio devices. See pa_list_devices().

Args

x: int

Number of the audio input and output devices.

setInputDevice(x)

Set the audio input device number. See pa_list_devices().

Args

x: int

Number of the audio device listed by Portaudio.

setOutputDevice(x)

Set the audio output device number. See pa_list_devices().

Args

x: int

Number of the audio device listed by Portaudio.

setInputOffset(x)

Set the first physical input channel.

Channel number x from the soundcard will be assigned to server’s channel one, channel number x + 1 to server’s channel two and so on.

Args

x: int

Channel number.

setOutputOffset(x)

Set the first physical output channel.

Server’s channel one will be assigned to soundcard’s channel number x, server’s channel two will be assigned to soundcard’s channel number x + 1 and so on.

Args

x: int

Channel number.

setInOutOffset(x)

Set the first physical input and output channels.

Set both offsets to the same value. See setInputOffset and setOutputOffset documentation for more details.

Args

x: int

Channel number.

setMidiInputDevice(x)

Set the Midi input device number. See pm_list_devices().

A number greater than the highest portmidi device index will opened all available input devices.

Args

x: int

Number of the Midi device listed by Portmidi.

setMidiOutputDevice(x)

Set the Midi output device number. See pm_list_devices().

Args

x: int

Number of the Midi device listed by Portmidi.

allowMicrosoftMidiDevices()

Allows the Microsoft Midi Mapper or GS Wavetable Synth device.

These are off by default because they crash on some systems.

setSamplingRate(x)

Set the sampling rate used by the server.

Args

x: int

New sampling rate, must be supported by the soundcard.

setBufferSize(x)

Set the buffer size used by the server.

Args

x: int

New buffer size.

setNchnls(x)

Set the number of output (and input if ichnls = None) channels used by the server.

Args

x: int

New number of channels.

setIchnls(x)

Set the number of input channels (if different of output channels) used by the server.

Args

x: int

New number of input channels.

setDuplex(x)

Set the duplex mode used by the server.

Args

x: int {0 or 1}

New mode. 0 is output only, 1 is both ways.

setVerbosity(x)

Set the server’s verbosity.

Args

x: int

A sum of values to display different levels:

• 1 = error

• 2 = message

• 4 = warning

• 8 = debug

setGlobalDur(x)

Set the global object duration (time to wait before stopping the object).

This value, if not 0, will override the dur argument of object’s play() and out() methods.

Args

x: float

New global duration.

setGlobalDel(x)

Set the global object delay time (time to wait before activating the object).

This value, if not 0, will override the del argument of object’s play() and out() methods.

Args

x: float

New global delay time.

setJackAuto(xin=True, xout=True)

Tells the server to auto-connect (or not) Jack ports to System ports.

Args

xin: boolean

Input Auto-connection switch. True is enabled (default) and False is disabled.

xout: boolean

Output Auto-connection switch. True is enabled (default) and False is disabled.

setJackAutoConnectInputPorts(ports)

Tells the server to auto-connect Jack input ports to pre-defined Jack ports.

Args

ports: list of list of strings

Name of the Jack ports to auto-connect to pyo input channels. There must be exactly one list of port(s) for each pyo input channel.

[[‘ports’, ‘to’, ‘channel’, ‘1’], [‘ports’, ‘to’, ‘channel’, ‘2’], …]

setJackAutoConnectOutputPorts(ports)

Tells the server to auto-connect Jack output ports to pre-defined Jack ports.

Args

ports: list of list of strings

Name of the Jack ports to auto-connect to pyo output channels. There must be exactly one list of port(s) for each pyo output channel.

[[‘ports’, ‘to’, ‘channel’, ‘1’], [‘ports’, ‘to’, ‘channel’, ‘2’], …]

setJackInputPortNames(name)

Change the short name of pyo’s input ports for the jack server.

This method must be called after the server is booted.

Args

name: string or list of strings

New name of input ports for the jack server. If name is a string, ‘_xxx’ (where xxx is the channel number) will be added to it for each input channel. If name is a list of strings, They will be used as is and there must be one for each input channel.

setJackOutputPortNames(name)

Change the short name of pyo’s output ports for the jack server.

This method must be called after the server is booted.

Args

name: string or list of strings

New name of output ports for the jack server. If name is a string, ‘_xxx’ (where xxx is the channel number) will be added to it for each output channel. If name is a list of strings, They will be used as is and there must be one for each output channel.

setJackAutoConnectMidiInputPort(ports)

Tells the server to auto-connect Jack midi input port to pre-defined Jack ports.

Args

ports: string or list of strings

Name of the Jack ports to auto-connect to pyo midi input channel.

[‘ports’, ‘to’, ‘midi’, ‘input’, …]

setJackAutoConnectMidiOutputPort(ports)

Tells the server to auto-connect Jack midi output port to pre-defined Jack ports.

Args

ports: string or list of strings

Name of the Jack ports to auto-connect to pyo midi output channel.

[‘ports’, ‘to’, ‘midi’, ‘output’, …]

setJackMidiInputPortName(name)

Change the short name of pyo’s midi input port for the jack server.

This method must be called after the server is booted.

Args

name: string

New name of midi input port for the jack server.

setJackMidiOutputPortName(name)

Change the short name of pyo’s midi output port for the jack server.

This method must be called after the server is booted.

Args

name: string

New name of midi output port for the jack server.

setIsJackTransportSlave(x)

Set if pyo’s server is slave to jack transport or not.

This method must be called before booting the server.

Args

x: boolean

If True, the server’s start and stop command will be slave to Jack transport. If False (the default) jack transport is ignored.

setGlobalSeed(x)

Set the server’s global seed used by random objects.

Args

x: int

A positive integer that will be used as the seed by random objects.

If zero, randoms will be seeded with the system clock current value.

setStartOffset(x)

Set the server’s starting time offset. First x seconds will be rendered offline as fast as possible.

Args

x: float

Starting time of the real-time processing.

setAmp(x)

Set the overall amplitude.

Args

x: float

New amplitude.

beginResamplingBlock(x)

Starts a resampling block.

This factor must be a power-of-two. A positive value means upsampling and a negative value means downsampling. After this call, every PyoObject will be created with an internal sampling rate and buffer size relative to the resampling factor. The method endResamplingBlock() should be called at the end of the code block using the resampling factor.

The Resample object can be used inside the resampling block to perform up or down resampling of audio signal created before the block.

Args

x: int, power-of-two

Resampling factor. Must be a power-of-two. A positive value starts an upsampling block while a negative value starts a downsampling block.

endResamplingBlock()

Ends a resampling block.

This call ends a code block using a sample rate different from the current sampling rate of the system.

The Resample object can be used after the resampling blick to perform up or down resampling of audio signal created inside the block.

shutdown()

Shut down and clear the server. This method will erase all objects from the callback loop. This method need to be called before changing server’s parameters like samplingrate, buffersize, nchnls, …

boot(newBuffer=True)

Boot the server. Must be called before defining any signal processing chain. Server’s parameters like samplingrate, buffersize or nchnls will be effective after a call to this method.

Args

newBuffer: bool

Specify if the buffers need to be allocated or not. Useful to limit the allocation of new buffers when the buffer size hasn’t change.

Therefore, this is useful to limit calls to the Python interpreter to get the buffers addresses when using Pyo inside a C/C++ application with the embedded server.

Defaults to True.

start()

Start the audio callback loop and begin processing.

stop()

Stop the audio callback loop.

recordOptions(dur=-1, filename=None, fileformat=0, sampletype=0, quality=0.4)

Sets options for soundfile created by offline rendering or global recording.

Args

dur: float

Duration, in seconds, of the recorded file. Only used by offline rendering. Must be positive. Defaults to -1.

filename: string

Full path of the file to create. If None, a file called pyo_rec.wav will be created in the user’s home directory. Defaults to None.

fileformat: int, optional

Format type of the audio file. This function will first try to set the format from the filename extension.

If it’s not possible, it uses the fileformat parameter. Supported formats are:

0. WAV - Microsoft WAV format (little endian) {.wav, .wave} (default)

1. AIFF - Apple/SGI AIFF format (big endian) {.aif, .aiff}

2. AU - Sun/NeXT AU format (big endian) {.au}

3. RAW - RAW PCM data {no extension}

4. SD2 - Sound Designer 2 {.sd2}

5. FLAC - FLAC lossless file format {.flac}

6. CAF - Core Audio File format {.caf}

7. OGG - Xiph OGG container {.ogg}

sampletype: int, optional

Bit depth encoding of the audio file.

SD2 and FLAC only support 16 or 24 bit int. Supported types are:

0. 16 bits int (default)

1. 24 bits int

2. 32 bits int

3. 32 bits float

4. 64 bits float

5. U-Law encoded

6. A-Law encoded

quality: float, optional

The encoding quality value, between 0.0 (lowest quality) and 1.0 (highest quality). This argument has an effect only with FLAC and OGG compressed formats. Defaults to 0.4.

recstart(filename=None)

Begins a default recording of the sound that is sent to the soundcard. This will create a file called pyo_rec.wav in the user’s home directory if no path is supplied or defined with recordOptions method. Uses file format and sample type defined with recordOptions method.

Args

filename: string, optional

Name of the file to be created. Defaults to None.

recstop()

Stop the previously started recording.

noteout(pitch, velocity, channel=0, timestamp=0)

Send a MIDI note message to the selected midi output device.

Arguments can be list of values to generate multiple events in one call.

Args

pitch: int

Midi pitch, between 0 and 127.

velocity: int

Amplitude of the note, between 0 and 127. A note with a velocity of 0 is equivalent to a note off.

channel: int, optional

The Midi channel, between 1 and 16, on which the note is sent. A channel of 0 means all channels. Defaults to 0.

timestamp: int, optional

The delay time, in milliseconds, before the message is sent to the output midi stream. A value of 0 means to play the note now. Defaults to 0.

afterout(pitch, velocity, channel=0, timestamp=0)

Send an aftertouch message to the selected midi output device.

Arguments can be list of values to generate multiple events in one call.

Args

pitch: int

Midi key pressed down, between 0 and 127.

velocity: int

Velocity of the pressure, between 0 and 127.

channel: int, optional

The Midi channel, between 1 and 16, on which the note is sent. A channel of 0 means all channels. Defaults to 0.

timestamp: int, optional

The delay time, in milliseconds, before the message is sent to the output midi stream. A value of 0 means to play the note now. Defaults to 0.

ctlout(ctlnum, value, channel=0, timestamp=0)

Send a control change message to the selected midi output device.

Arguments can be list of values to generate multiple events in one call.

Args

ctlnum: int

Controller number, between 0 and 127.

value: int

Value of the controller, between 0 and 127.

channel: int, optional

The Midi channel, between 1 and 16, on which the message is sent. A channel of 0 means all channels. Defaults to 0.

timestamp: int, optional

The delay time, in milliseconds, before the message is sent to the output midi stream. A value of 0 means to play the note now. Defaults to 0.

programout(value, channel=0, timestamp=0)

Send a program change message to the selected midi output device.

Arguments can be list of values to generate multiple events in one call.

Args

value: int

New program number, between 0 and 127.

channel: int, optional

The Midi channel, between 1 and 16, on which the message is sent. A channel of 0 means all channels. Defaults to 0.

timestamp: int, optional

The delay time, in milliseconds, before the message is sent to the output midi stream. A value of 0 means to play the note now. Defaults to 0.

pressout(value, channel=0, timestamp=0)

Send a channel pressure message to the selected midi output device.

Arguments can be list of values to generate multiple events in one call.

Args

value: int

Single greatest pressure value, between 0 and 127.

channel: int, optional

The Midi channel, between 1 and 16, on which the message is sent. A channel of 0 means all channels. Defaults to 0.

timestamp: int, optional

The delay time, in milliseconds, before the message is sent to the output midi stream. A value of 0 means to play the note now. Defaults to 0.

bendout(value, channel=0, timestamp=0)

Send a pitch bend message to the selected midi output device.

Arguments can be list of values to generate multiple events in one call.

Args

value: int

14 bits pitch bend value. 8192 is where there is no bending, 0 is full down and 16383 is full up bending.

channel: int, optional

The Midi channel, between 1 and 16, on which the message is sent. A channel of 0 means all channels. Defaults to 0.

timestamp: int, optional

The delay time, in milliseconds, before the message is sent to the output midi stream. A value of 0 means to play the note now. Defaults to 0.

sysexout(msg, timestamp=0)

Send a system exclusive message to the selected midi output device.

Arguments can be list of values/messages to generate multiple events in one call. Implemented only for portmidi, this method is not available when using jack as the midi backend.

Args

msg: str

A valid system exclusive message as a string. The first byte must be 0xf0 and the last one must be 0xf7.

timestamp: int, optional

The delay time, in milliseconds, before the message is sent to the output midi stream. A value of 0 means to play the message now. Defaults to 0.

makenote(pitch, velocity, duration, channel=0)

Send MIDI noteon/noteoff messages to the selected midi output device.

This method will send a noteon message to the selected midi output device and after a delay of duration milliseconds, it will send the corresponding noteoff message.

Arguments can be list of values to generate multiple events in one call.

Args

pitch: int

Midi pitch, between 0 and 127.

velocity: int

Amplitude of the note, between 0 and 127. A note with a velocity of 0 is equivalent to a note off.

duration: int

The delay time, in milliseconds, before the noteoff message is sent to the output midi stream.

channel: int, optional

The Midi channel, between 1 and 16, on which the note is sent. A channel of 0 means all channels. Defaults to 0.

addMidiEvent(status, data1=0, data2=0)

Add a MIDI event in the server processing loop.

This method can be used to programmatically simulate incoming MIDI events. In an embedded framework (ie. pyo inside puredata, openframeworks, etc.), this is useful to control a MIDI-driven script from the host program. Arguments can be list of values to generate multiple events in one call.

The MIDI event buffer is emptied at the end of each processing block. So, for events to be processed, addMidiEvent should be called at the beginning of the block. If you use audio objects to generate MIDI events, they should be created before the rest of the processing chain.

Args

status: int

The status byte, indicating the type of event and the MIDI channel. Typical event type are:

• 128 -> 143: Noteoff

• 144 -> 159: Noteon

• 176 -> 191: Control change

• 192 -> 207: Program change

• 208 -> 223: After touch

• 224 -> 239: Pitch bend

data1: int, optional

The first data byte (pitch for a midi note, controller number for a control change). Defaults to 0.

data2: int, optional

The second data byte (velocity for a midi note, value for a control change). Defaults to 0.

getSamplingRate()

Return the current sampling rate.

getNchnls()

Return the current number of channels.

getBufferSize()

Return the current buffer size.

getGlobalDur()

Return the current global duration.

getGlobalDel()

Return the current global delay time.

getGlobalSeed()

Return the current global seed.

getIsStarted()

Returns 1 if the server is started, otherwise returns 0.

getIsBooted()

Returns 1 if the server is booted, otherwise returns 0.

deactivateMidi()

Deactivate Midi callback. Must be called before the boot() method.

getMidiActive()

Returns 1 if Midi callback is active, otherwise returns 0.

getStreams()

Returns the list of Stream objects currently in the Server memory.

getNumberOfStreams()

Returns the number of streams currently in the Server memory.

setServer()

Sets this server as the one to use for new objects when using the embedded device.

getInputAddr()

Return the address of the input buffer.

getOutputAddr()

Return the address of the output buffer.

getServerID()

Return the server ID.

getServerAddr()

Return the address of the server.

getEmbedICallbackAddr()

Return the address of the interleaved embedded callback function.

getCurrentTime()

Return the current time as a formatted string.

getCurrentTimeInSamples()

Return the current time in number of elapsed samples since the server was started.

getCurrentAmp()

Return the current amplitudes as a tuple of nchnls length.

setAutoStartChildren(state)

Giving True to this method tells pyo that a call to the play, out or stop method of audio objects should propagate to the other audio objects given as arguments. This can be used to control an entire dsp chain just by calling methods on the very last object.

With setAutoStartChildren(True), a call to the out method will automatically triggers the play method of the Fader object given as mul argument.

>>> a = RCOsc(freq=150, mul=Fader(1, 1, mul=.3)).out()

With setAutoStartChildren(True), a call to the stop method will also propagate to audio objects given as arguments. The stop method has an argument wait, useful to postpone the moment when to really stop the process. The waiting value is also propagated to the other audio objects, but not for the objects assigned to a `mul argument. This property allows to do things like the next snippet. On a call a.stop(1) , while the Linseg assigned to the frequency waits for 1 second before actually stopping its process, the Fader assigned to the mul argument starts its fadeout immediately.

>>> freq = Linseg([(0,500),(1, 750),(2, 500)], loop=True)
>>> a = Sine(freq=freq, mul=Fader(1, 1, mul=.2)).out()

Sometime, we still want the process assigned to a mul attribute to wait before stopping its process. See the next case. The amplitude envelope is itself modulated in amplitude by another envelope. We still want the fadeout to start immediately, but its own envelope has to wait, otherwise, the amplitude will cut off before having the time to complete the fadeout. To do this, we call useWaitTimeOnStop() on the object we want to force to wait.

>>> lf = Linseg([(0,0),(0.2,0.1),(0.4,0)], loop=True)
>>> lf.useWaitTimeOnStop()
>>> freq = Linseg([(0,500),(1, 750),(2, 500)], loop=True)
>>> a = Sine(freq=freq, mul=Fader(1, 1, mul=lf)).out()

Note

The waiting time doesn’t propagate to objects used in audio arithmetic. Code like this (for which you want to call out.stop()):

>>> a = Sine(500)
>>> a.setStopDelay(1)
>>> b = Sine(750)
>>> b.setStopDelay(2)
>>> out = Freeverb(a+b, size=0.8, mul=0.3).out()
>>> out.setStopDelay(4)

Should be written as:

>>> a = Sine(500)
>>> a.setStopDelay(1)
>>> b = Sine(750)
>>> b.setStopDelay(2)
>>> c = Mix([a, b], voices=1)
>>> c.setStopDelay(2)
>>> out = Freeverb(c, size=0.8, mul=0.3).out()
>>> out.setStopDelay(4)

process()

Tell the server to compute a single buffer size of audio samples.

The audio backend of the server must be set to manual to use this method. It’s probably only useful in the context of internal testing.

property amp

float. Overall amplitude.

property startoffset

float. Starting time of the real-time processing.

property verbosity

int. Server verbosity.

property globalseed

int. Server global seed.