API

REW has a REST API accessible over http at localhost (127.0.0.1), the default port is 4735. It cannot be accessed outside the machine REW is running on. To start the API server use the button on the API preferences or run REW with the -api argument. On Windows with the default REW installation that could be done using

C:\Program Files\REW\roomeqwizard.exe -api

On macOS it could be done from a terminal using

open -a REW.app --args -api

To specify a different port, e.g. 4567, add -port 4567. Port numbers below 1024 will be ignored. The default port will be assumed in the examples which follow. To run REW without a GUI use the -nogui argument, but if you do that you must use the application shutdown command to close REW when done.

All API GET methods are available whenever the API is running, allowing the API to be used to retrieve data from REW. PUT and POST are also supported by default for most endpoints, but to control REW via the API to make automated sweep measurements requires a Pro upgrade license.

The API documentation is served by swagger-ui and can be accessed by browsing to localhost:4735. The OpenAPI specification for the API can be accessed at localhost:4735/doc.json or localhost:4735/doc.yaml for JSON or YAML formats. Setting values using the data models can be done with POST or, for models with multiple fields, PUT, however both methods will accept data models with a subset of the defined fields.

Array encoding

Arrays are transferred as Base64-encoded strings generated from the raw bytes of the 32-bit float sample values. Base64 decoding is well supported in programming languages, here is how to recover the arrays in Java:

    public static float[] decodeArray(String base64Encoded) {
        return byteToFloatArray(Base64.getDecoder().decode(base64Encoded));
    }

    private static float[] byteToFloatArray(byte[] bytes) {
        FloatBuffer buf = ByteBuffer.wrap(bytes).asFloatBuffer();
        float[] floatArray = new float[buf.limit()];
        buf.get(floatArray);
        return floatArray;
    }

Application

The /application endpoint provides application-level control of REW. A GET at the /application/commands endpoint returns the list of commands, to issue a command POST it to the /application/command endpoint. There is a shutdown command for use when REW is run without a GUI. There is a "Clear command in progress" command to reset the API's internal record of a command that has been previously accepted by an endpoint. That record is used to prevent commands being run before a previous command has completed, the "Clear command in progress" is a fallback in case an error has prevented a command from completing and clearing the record.

Any errors that have been logged can be read from the /application/errors endpoint, that returns a list of ErrorMessage structures. The most recent error can be read from /application/last-error. Rather than polling for errors with GET a subscription can be added to be notified if an error occurs. That is done by making a POST to the /application/errors/subscribe endpoint. The body contains a Subscriber object which has a URL string and an optional set of parameters, there are no parameters for error subscriptions. For example, to have errors posted to a /errors endpoint a subscription could be:

{
  "url": "http://127.0.0.1:5374/errors"
}

If REW does not get an OK response to an update the subscription will be cancelled. It can also be removed by posting the same Subscriber data to the /application/errors/unsubscribe endpoint. The list of subscribers can be read from /application/errors/subscribers.

Any warnings that have been logged can be read from the /application/warnings endpoint, that returns a list of WarningMessage structures. The most recent warning can be read from /application/last-warning. Rather than polling for warnings with GET a subscription can be added to be notified if an warning occurs. That is done by making a POST to the /application/warnings/subscribe endpoint. The body contains a Subscriber object which has a URL string and an optional set of parameters, there are no parameters for warning subscriptions. For example, to have warnings posted to a /warnings endpoint a subscription could be:

{
  "url": "http://127.0.0.1:5374/warnings"
}

If REW does not get an OK response to an update the subscription will be cancelled. It can also be removed by posting the same Subscriber data to the /application/warnings/unsubscribe endpoint. The list of subscribers can be read from /application/warnings/subscribers.

Audio

The /audio endpoint provides full control of REW's audio input and output selections and associated controls. A GET at the audio endpoint returns its current status, including whether audio is enabled (true unless the -noaudio command line option was used) and whether the audio endpoints are ready for use, which becomes true a few seconds after REW starts up.

On Windows REW has a choice of Java or ASIO audio drivers, on other platforms Java is the only choice. The /audio/driver endpoint allows the driver to be read (GET) or set (POST). The driver choices are available at /audio/driver-types. All Java-related items are under /audio/java/, ASIO-related items are under /audio/asio/. Items under those paths are only accessible when the respective driver has been selected.

The sample rate can be read and set at the /audio/samplerate endpoint. When setting sample rate the unit can be Hz, kHz or absent. If absent Hz is assumed. If the current audio interface cannot support the requested sample rate a 400 Bad Request status will be returned with a list of the rates that are available for selection. The sample rates supported for the current interface selection can be read from /audio/samplerates.

The global option to treat 32-bit data as 24-bit can be read and set at the /audio/configuration endpoint.

The current input calibration configuration, including any mic cal file(s) being used, can be read at the /audio/input-cal endpoint. This endpoint only supports PUT to make changes since the current input selection the cal data relates to is determined by the driver, input device and input selections. Although the current input selection field in the returned InputCalConfiguration structure is read-only it is recommended that it be included when changing settings, REW will check that it still corresponds to the input in use and reject the request if the input has changed. To clear a cal file set its file path to an empty string. Note that backslashes in file paths must be escaped (meaning a double backslash is required), or forward slashes may be used.

The current output ("soundcard") calibration configuration, including any cal file being used, can be read at the /audio/output-cal endpoint. This endpoint only supports PUT to make changes since the current output selection the cal data relates to is determined by the driver, output device and output selections. Although the current output selection field in the returned OutputCalConfiguration structure is read-only it is recommended that it be included when changing settings, REW will check that it still corresponds to the output in use and reject the request if the output has changed. To clear a cal file set its file path to an empty string. Note that backslashes in file paths must be escaped (meaning a double backslash is required), or forward slashes may be used.

Java settings

All Java driver items are under /audio/java/. After selecting the Java driver the current input device, input, output device and output can be read or set at /audio/java/input-device, /audio/java/input, /audio/java/output-device and /audio/java/output. The lists of available devices, inputs and outputs to choose from are at /audio/java/input-devices, /audio/java/inputs, /audio/java/output-devices and /audio/java/outputs. The input and output are specific to their respective device, if the device is changed the list of inputs or outputs available will also change.

By default REW will select the first named (i.e. non default) input or output available when a device is selected, which for many devices will be the only option. After the device and input or output have been selected the input channel and output channel may be selected. If applicable (depending on the timing reference mode for measurement) the timing reference output channel and the input reference (loopback) channel may also be selected. If multiple inputs are being captured (Pro upgrade) the last input channel of the input range may be selected. If using Virtual balanced input mode the ref input channel is the second channel used to form the balanced pair. The channel endpoints are /audio/java/input-channel, /audio/java/ref-input-channel, /audio/java/last-input-channel, /audio/java/output-channel and /audio/java/ref-output-channel. Input channels are numeric, indexed from 1. The number of input channels can be read from /audio/java/num-input-channels. Output channels are strings and have a L+R option to select the front pair as output. The output channel choices can be read from /audio/java/output-channels, the ref output channel choices from /audio/java/ref-output-channels.

REW supports up to 16 java output channels. Those channels can be mapped to any 16 of the hardware channels on an interface through the /audio/java/output-channel-mapping endpoint. A GET returns the list of the current mappings, the length of the list will be the number of output channels on the currently selected output device up to a maximum of 16. The mappings can be changed by POST of a mapping list to the endpoint. If an empty string is provided as the channel label the hardware channel will be used as the label. If the list posted has more channel entries than the output device supports any mappings outside the range of the output device hardware channel count will be ignored.

The Java driver can be restricted to only accesses the audio interface in stereo by setting the stereoOnly flag at /audio/java/stereo-only. Note that on Windows only the WASAPI-exclusive Java device entries (names starting "EXCL:") support multichannel operation, other entries only offer mono or stereo connections to the interface.

ASIO settings

All ASIO driver items are under /audio/asio/. After selecting the ASIO driver the current device, input and output can be read or set at /audio/asio/device /audio/asio/input and /audio/asio/output. The lists of available devices, inputs and outputs to choose from are at /audio/asio/devices, /audio/asio/inputs, and /audio/asio/outputs. The input and output are specific to their respective device, if the device is changed the list of inputs or outputs available will also change.

If applicable (depending on the timing reference mode for measurement) the timing reference output channel and the input reference (loopback) channel may also be selected. If multiple inputs are being captured (Pro upgrade) the last input of the input range may be selected. If using Virtual balanced input mode the ref input is the second channel used to form the balanced pair. A second ASIO output can be selected by first enabling it at /audio/asio/secondary-output-enable then selecting it at /audio/asio/secondary-output.

A forced reload of the ASIO driver can be triggered by posting a reload command to /audio/asio/command.

Input levels

The /input-levels endpoint provides access to the current input level via subscription whenever the inputs are being monitored. Monitoring is started or stopped by posting a command to the /input-levels/command endpoint. The list of commands is retrieved from the The /input-levels/commands endpoint. Sending a command to start monitoring input levels will start audio capture if it is not already running. Audio capture will continue until a command is sent to stop monitoring input levels.

The last input levels update can be read from the /input-levels/last-levels endpoint. That will return the last InputLevels object generated by an input levels update. More commonly levels would be monitored through a subscription.

A subscription can be added by making a POST to the /input-levels/subscribe endpoint. The body contains a Subscriber object which has a URL string and an optional set of parameters, input levels subscriptions accept a "unit" parameter that specifies the units of the input level values. The accepted units can be read from the /input-levels/units endpoint. If the unit is omitted levels are returned as dBFS. To have input level updates posted to a /input-levels endpoint a subscription could be:

{
  "url": "http://127.0.0.1:5374/input-levels",
  "parameters": {"unit": "dBV"}
}

If REW does not get an OK response to an update the subscription will be cancelled. It can also be removed by posting the same Subscriber data to the /input-levels/unsubscribe endpoint. The list of subscribers can be read from /input-levels/subscribers. Levels are returned as an InputLevels object which has the unit, arrays of the rms and peak levels for each input since the last update and a timeSpanSeconds value with the time period over which the values were calculated.

Measurements

The /measurements endpoint provides access to any measurements REW has generated or loaded. A GET at the /measurements endpoint returns a summary list of the current measurements. A GET at the /measurements/:id gets the summary for the measurement at index id, indices start from 1. A DELETE at the /measurements/:id endpoint deletes that measurement, DELETE at the /measurements endpoint deletes all the measurements. N.B. There are no delete confirmations, use with care!

A GET at the /measurements/commands endpoint returns a list of commands, to issue a command POST it to the /measurements/command endpoint. The "Save all" command has a parameter for the path to the file to be saved to, if the file exists it will be overwritten. A note to be saved with the measurement can optionally be specified as a second parameter. The "Load" command has a list of the filenames to load as parameters in the command. Note that if a file path has backslash as the path separator the string for the path will need to escape the backslash entries, i.e. use a double backslash instead or replace backslash by forward slash.

A measurement's name and/or notes can be changed by a PUT at the /measurements/:id endpoint, putting a MeasurementSummary object which has the title and/or notes field set to the new value.

An individual measurement can be processed by POST of a ProcessSingleMeasurement object to the /measurements/:id/command endpoint. The list of commands for individual measurements can be retrieved from the /measurements/:id/commands endpoint. They include commands to Save, generate a minimum phase version, generate an excess phase version, generate a Mic in box correction, make a response copy, merge cal data to IR, Trim IR to windows, smooth, generate waterfalls, generate spectrograms, Estimate IR delay, offset t=0 and add SPL offset. Some of those commands produce a new measurement (generate a minimum phase version, generate an excess phase version, generate a Mic in box correction, make a response copy, merge cal data to IR, Trim IR to windows), others make changes to the measurement itself.

The commands may have associated parameters, for example the save command requires a "filename" parameter with the filename to save to, smooth requires a "smoothing" setting. Note that if a file path has backslash as the path separator the string for the path will need to escape the backslash entries, i.e. use a double backslash instead or replace backslash by forward slash. Here is an example of a save command:

{
  "command": "Save",
  "parameters": {filename: "c:/users/myusername/downloads/myfile.mdat"}
}

The command has an optional resultUrl, if included the result of the process will be posted to the URL. The process result can be read from /measurements/process-measurements-result but may not be available immediately, some processes take time to complete. The ProcessMeasurementsResult object has the name of the process it relates to, a message (which may be absent or empty) and a set of results for each measurement that was processed which are returned as a map of result names and result values, all strings. For individual measurements there will be one map entry.

Generate waterfall and generate equalised waterfall require a "mode" parameter ("Fourier" or "Burst decay"), a parameter "slices" for the number of slices to generate and a set of additional parameters the depend on the mode. Fourier requires "left window type", "right window type", "window width ms", "time range ms", "rise time ms" and "use csd mode" parameters. Burst decay requires "bandwidth" and "periods" parameters. Here are some examples of commands to generate a waterfall:

{
  "command": "Generate waterfall",
  "parameters": {
    "mode": "Fourier",
    "slices": "101",
    "left window type": "Hann",
    "right window type": "Tukey 0.25",
    "window width ms": "300",
    "time range ms": "500",
    "rise time ms": "150",
    "use csd mode": "false",
    "ppo": "48",
    "smoothing": "1/48"},
  "resultUrl": "http://127.0.0.1:5374/waterfalls"
}

{
  "command": "Generate waterfall",
  "parameters": {
    "mode": "Burst decay",
    "slices": "101",
    "periods": "30",
    "ppo": "48",
    "bandwidth": "1/3"},
  "resultUrl": "http://127.0.0.1:5374/waterfalls"
}

Generate spectrogram also requires a "mode" parameter ("Fourier", "Wavelet, "Airy CWT, "Morlet CWT, "Burst decay"), a parameter "slices" for the number of slices to generate, a "ppo" parameter for the frequency resolution of the result, an "amplitude" parameter for the data values and a set of additional parameters the depend on the mode. Fourier requires "window type", "window width ms", "before ms" and "after ms" parameters. Burst decay requires "bandwidth" and "periods" parameters. The wavelet modes require "bandwidth", "before ms" and "after ms" parameters. Here are some examples of commands to generate a spectrogram:

{
  "command": "Generate spectrogram",
  "parameters": {
    "mode": "Fourier",
    "slices": "551",
    "amplitude": "Log (dB SPL)",
    "window type": "Gaussian",
    "window width ms": "300",
    "before ms": "50",
    "after ms": "500",
    "ppo": "48"},
  "resultUrl": "http://127.0.0.1:5374/spectrograms"
}

{
  "command": "Generate spectrogram",
  "parameters": {
    "mode": "Morlet CWT",
    "slices": "551",
    "amplitude": "Linear (% peak)",
    "before ms": "5",
    "after ms": "50",
    "bandwidth": "1/3",
    "ppo": "48"},
  "resultUrl": "http://127.0.0.1:5374/spectrograms"
}

The generate waterfall and spectrogram commands return the 2D data in a ProcessResult. The result includes an array of the frequencies, an array of the times (or periods for Burst Decay) and a set of arrays of SPL data vs frequency for each time index. It is best to specify a resultUrl for those as the generation can take some time. Note that the commands do not affect the REW GUI and will not produce a result in the GUI, they are provided as a way to generate and extract data.

Minimum phase version and Excess phase version require parameters for "include cal", "append lf tail" and "append hf tail" which are boolean values. If "append lf tail" is true the parameters must include an "lf tail start" frequency in Hz and an "lf tail slope" in dB/octave, which must be ≥ 0. If "append hf tail" is true the parameters must include an "hf tail start" frequency in Hz and an "hf tail slope" in dB/octave, which must be ≤ 0, along with a "frequency warping" boolean. If either "append lf tail" or "append hf tail" are false a "replicate data" boolean is required. Here is an example of a command to generate a minimum phase version

{
  "command": "Minimum phase version",
  "parameters": {
    "include cal": "true",
    "append lf tail": "true",
    "lf tail start": "27",
    "lf tail slope": "12.0",
    "append hf tail": "true",
    "hf tail start": "15000",
    "hf tail slope": "-18.0",
    "frequency warping": "false",
    "replicate data": "false"},
  "resultUrl": "http://127.0.0.1:5374/minphase"
}

Smooth requires a smoothing parameter:

{
  "command": "Smooth",
  "parameters": {
    "smoothing": "1/3"
  }
}

Offset t=0 requires parameters for "offset" and "unit", where unit is for the offset value and may be seconds, metres, feet or samples. Here is an example of a command to offset t=0

{
  "command": "Offset t=0",
  "parameters": {
    "offset": "0.5",
    "unit": "metres"},
  "resultUrl": "http://127.0.0.1:5374/result"
}

Add SPL offset requires an offset parameter:

{
  "command": "Add SPL offset",
  "parameters": {
    "offset": -3.2
  }
}

A GET at the /measurements/selected endpoint returns the index of the currently selected measurement. To select a measurement POST its index to /measurements/selected.

A subscription can be added to be notified of any changes that are made to the list of measurements, including measurements being added or deleted, the contents of a measurement changing, or a measurement becoming selected. That is done by making a POST to the /measurements/subscribe endpoint. The body contains a Subscriber object which has a URL string and an optional set of parameters, there are no parameters for measurement subscriptions. For example, to have changes posted to a /measurements endpoint a subscription could be:

{
  "url": "http://127.0.0.1:5374/measurements"
}

If REW does not get an OK response to an update the subscription will be cancelled. It can also be removed by posting the same Subscriber data to the /measurements/unsubscribe endpoint. The list of subscribers can be read from /measurements/subscribers. The change notifications are conveyed in a MeasurementsListChange object, which has a string with a description of the type of change and indices for the range of measurements the change affects.

A set of measurements can be processed by POST of a ProcessMeasurements object to the /measurements/process-measurements endpoint. The list of process commands can be read from /measurements/process-commands. The command parameters will vary according to the process to be carried out, see details below. The ProcessMeasurements object has a list of the indices of the measurements to be processed. It also has an optional URL, if included the result of the process will be posted to the URL. The process result can be read from /measurements/process-result but may not be available immediately, some processes take time to complete. The ProcessMeasurementsResult object has the name of the process it relates to, a message (which may be absent or empty) and a set of results for each measurement that was processed which are returned as a map of result names and result values, all strings.

The Align SPL process has these parameters:

targetdB: either "average" to align to the average SPL of the measurements or an SPL value in dB, e.g. "75.0"
frequencyHz: the centre frequency for the alignment in Hz, e.g. "1000"
spanOctaves: the alignment span in octaves, e.g. "2"

Here is an example of a ProcessMeasurements object to align SPL:

{
  "processName": "Align SPL",
  "measurementIndices": [1,3,4],
  "parameters": {"targetdB": "85.0", "frequencyHz": "1000", "spanOctaves": 2},
  "resultUrl": ""
}

The results of Align SPL are available as soon as the process command is acknowledged.

The Arithmetic process operates on pairs of measurements, the first measurement is designated A, the second B. Arithmetic has these parameters, function is mandatory but others are optional if applicable to the specified function:

function: The arithmetic function to perform, the list of functions can be read from /measurements/arithmetic-functions
regularisation: Percentage regularisation for division and inversion functions
lowerLimit: the lower frequency limit in Hz for band limited functions (division or inversion)
upperLimit: the lower frequency limit in Hz for band limited functions (division or inversion)
mergeFrequency: the frequency at which to perform a merge function
mergeBlend: true if a merge function is to be blended over a span
targetLevel: target level in dB for an inversion function
autoTarget: true if target level is to be set automatically for an inversion function
excludeNotches: true if notches are to be excluded for an inversion function

Here is an example of a ProcessMeasurements object to perform A * B trace arithmetic:

{
  "processName": "Arithmetic",
  "measurementIndices": [1,3],
  "parameters": {"function": "A * B"},
  "resultUrl": ""
}

Here is an example of a ProcessMeasurements object to perform 1 / A trace arithmetic:

{
  "processName": "Arithmetic",
  "measurementIndices": [1,2],
  "parameters": {
    "function": "1 / A",
    "regularisation": "0.5",
    "lowerLimit": "100",
    "upperLimit": "1000",
    "targetLevel": "75.0",
    "autoTarget": "false",
    "excludeNotches": "true"
  },
  "resultUrl": ""
}

The frequency response for a measurement can be obtained by a GET at /measurements/:id/frequency-response. A FrequencyResponse object is returned, with a smoothing setting, a start frequency, a points-per-octave figure (for log-spaced data) or a frequency step value (for linear-spaced data), and Base64-encoded strings of response magnitudes and phases generated from the raw bytes of the 32-bit float magnitude and phase values at each frequency. The default unit for the magnitude data is SPL but other units can be requested by including a unit query value in the url, e.g. ?unit=dBFS. The list of units available can be read from /measurements/frequency-response/units. The default smoothing is whatever the measurement currently uses, but other smoothings can be requested by including a smoothing query value in the url, e.g. ?smoothing="1/12". The list of smoothing choices can be read from /measurements/frequency-response/smoothing-choices. Whether the returned data is log-spaced or linear-spaced depends on the measurement, but log-spaced data can be forced by including a ppo query value in the url, e.g. ?ppo=96. Note that to avoid sampling artefacts log-spaced data will be smoothed to ppo/2 if a greater smoothing has not already been applied.

The impulse response for a measurement (if it has one) can be obtained by a GET at /measurements/:id/impulse-response. An ImpulseResponse object is returned, with a start time, a description of the timing reference used, a sample interval in seconds, a sample rate in Hz and a Base64-encoded string of response data. The default unit for the data is Percent but other units can be requested by including a unit query value in the url, e.g. ?unit=dBFS. The list of units available can be read from /measurements/impulse-response/units. Impulse response queries can return the windowed portion of the data by including a query value ?windowed=true in the url. Data is normalised by default, for data that is not normalised include a query value ?normalised=false in the url.

The impulse response for a measurement's filters (if there are any) can be obtained by a GET at /measurements/:id/filters-impulse-response. An ImpulseResponse object is returned, with a start time, a sample interval in seconds, a sample rate in Hz and a Base64-encoded string of response data. The sample rate and length of the response must be specified by including query values in the url, e.g. ?samplerate=48000&length=65536. The maximum length accepted is 4,194,304 samples.

A measurement's equaliser selection can be read from the /measurements/:id/equaliser endpoint and changed by a POST to that endpoint. The available equalisers can be read from the /eq/equalisers endpoint.

A measurement's target settings can be read from the /measurements/:id/target-settings endpoint and changed by a POST or PUT to that endpoint.

A measurement's target level can be read from the /measurements/:id/target-level endpoint and changed by a POST to that endpoint.

A measurement's room curve settings can be read from the /measurements/:id/room-curve-settings endpoint and changed by a POST or PUT to that endpoint.

A measurement's filters can be read from the /measurements/:id/filters endpoint. A list of FilterSetting objects is returned with the current settings for each filter of the measurement's equaliser. Note that the number and type of filters available depend on the equaliser. The settings for an individual filter can be changed by a PUT of a FilterSetting object to the endpoint. Multiple (or all) filters can be changed by a POST of a FilterList object to the endpoint.

A measurement's EQ target response can be read from the /measurements/:id/target-response endpoint. A FrequencyResponse object is returned, with a start frequency, a points-per-octave figure and a Base64-encoded string of response magnitudes. The default unit for the response data is SPL but other units can be requested by including a unit query value in the url, e.g. ?unit=dBFS. The list of units available can be read from /measurements/frequency-response/units. The data returned is log spaced at 96 PPO unless there is a ppo query value in the url, e.g. ?ppo=48.

The frequency response for the predicted result of a measurement's EQ filters can be obtained by a GET at /measurements/:id/eq/frequency-response. A FrequencyResponse object is returned, with a smoothing setting, a start frequency, a points-per-octave figure (for log-spaced data) or a frequency step value (for linear-spaced data), and Base64-encoded strings of response magnitudes and phases generated from the raw bytes of the 32-bit float magnitude and phase values at each frequency. The default unit for the magnitude data is SPL but other units can be requested by including a unit query value in the url, e.g. ?unit=dBFS. The list of units available can be read from /measurements/frequency-response/units. The default smoothing is whatever the measurement currently uses, but other smoothings can be requested by including a smoothing query value in the url, e.g. ?smoothing="1/12". The list of smoothing choices can be read from /measurements/frequency-response/smoothing-choices. Whether the returned data is log-spaced or linear-spaced depends on the measurement, but log-spaced data can be forced by including a ppo query value in the url, e.g. ?ppo=96. Note that to avoid sampling artefacts log-spaced data will be smoothed to ppo/2 if a greater smoothing has not already been applied.

The impulse response for for the predicted result of a measurement's EQ filters (if the measurement itself has an impulse response) can be obtained by a GET at /measurements/:id/eq/impulse-response. An ImpulseResponse object is returned, with a start time, a description of the timing reference used, a sample interval in seconds, a sample rate in Hz and a Base64-encoded string of response data. The default unit for the data is Percent but other units can be requested by including a unit query value in the url, e.g. ?unit=dBFS. The list of units available can be read from /measurements/impulse-response/units. Impulse response queries can return the windowed portion of the data by including a query value ?windowed=true in the url. Data is normalised by default, for data that is not normalised include a query value ?normalised=false in the url.

Measure

The /measure endpoint provides access to REW's measurement capabilities. A GET at the /measure/commands endpoint returns the list of commands, to issue a command POST it to the /measure/command endpoint. The commands allow different types of measurement (SPL, impedance, impedance calibration) to be made. There is also a command to cancel a measurement that is in progress.

The /measure/naming endpoint provides access to the naming settings. They can be changed by POST or PUT, the response will show the format of the next measurement name. The options for naming can be read from /measure/naming/naming-options. If a date or time is being added to the measurement name the formats for that can be read from /measure/naming/date-time-formats. The empty string format option "" uses the default date and time format for the machine.

The /measure/notes endpoint provides access to the notes that will be included in the next measurement made. They can be changed or set by POST.

The measurement level can be read and set at the /measure/level endpoint. A variety of units are supported, the list of accepted units is available at /measure/level/units. The default unit is dBFS, if a unit is not specified when setting level dBFS will be assumed.

The /measure/protection-options endpoint provides access to the options to abort measurements if heavy clipping is detected on the input or if the SPL exceeds a limit. They can be changed by POST or PUT.

The /measure/sweep/configuration endpoint provides access to the measurement sweep configuration, including start frequency, end frequency, length and whether to fill silence with dither. The configuration can be changed using POST or PUT.

The /measure/sweep/repetitions endpoint provides access to the number of sweep repetitions. The number can be changed using POST. Note that multiple sweeps cannot be used with USB mics or when using the acoustic or wired timing reference modes.

The /measure/timing endpoint provides access to settings for the timing reference. There are endpoints to get and set the reference (/measure/timing/reference) and further endpoints to configure the options for the chosen reference.

The /measure/timing-offset endpoint provides access to the timing offset to apply when using a timing reference. The value can be changed using POST.

The /measure/playback-mode endpoint provides access to the playback mode. The value can be changed using POST, the options can be read from /measure/playback-mode/choices.

The /measure/measurement-mode endpoint provides access to the measurement mode, for single, repeated or sequential measurements. The value can be changed using POST, the options can be read from /measure/measurement-mode/choices.

The /measure/number-of-repetitions endpoint provides access to the number of measurements that will be made when measuring SPL in repeated measurement mode with playback from REW. The value can be changed using POST.

The /measure/sequential-channels endpoint provides access to the list of channels that will be measured when measuring SPL in sequential measurement mode with playback from REW. The list can be changed using POST. The full list of channel choices can be read from /measure/sequential-choices.

The /measure/start-delay endpoint provides access to the delay in seconds before a measurement starts. The value can be changed using POST.

The /measure/invert-second-output endpoint provides access to whether to invert the second output when two outputs are driven. The value can be changed using POST.

The /measure/fill-silence-with-dither endpoint provides access to setting for whether to fill silent parts of the output with 16-bit dither to ensure the replay path is active. The value can be changed using POST.

The /measure/capture-noise-floor endpoint provides access to setting for whether to capture the noise floor before an SPL measurement. The value can be changed using POST.

A subscription can be added to be notified of progress during a measurement. That is done by making a POST to the /measure/subscribe endpoint. The body contains a Subscriber object which has a URL string and an optional set of parameters, there are no parameters for measure subscriptions. For example, to have changes posted to a /measurement-progress endpoint a subscription could be:

{
  "url": "http://127.0.0.1:5374/measurement-progress"
}

If REW does not get an OK response to an update the subscription will be cancelled. It can also be removed by posting the same Subscriber data to the /measure/unsubscribe endpoint. The list of subscribers can be read from /measure/subscribers. The change notifications are conveyed as a string.

Alignment tool

The /alignment-tool endpoint provides access to REW's alignment tool. See Alignment tool for details of the operation of the tool.

A GET at /alignment-tool/commands returns the list of commands the tool accepts. Use POST to send a command to /alignment-tool/command. The alignment mode is at the /alignment-tool/mode endpoint. The choice of modes can be read from /alignment-tool/modes. The frequency at which the alignment is performed is at the /alignment-tool/frequency endpoint. The alignment is performed on a pair of measurements A and B, the endpoints for the index of each of those measurements are /alignment-tool/index-a and /alignment-tool/index-b. The gain, delay and invert for each measurement have individual endpoints to GET them, they can be changed by POST. After carrying out an alignment the new delay in milliseconds for measurement B can be read from the /alignment-tool/delay-b endpoint. A FrequencyResponse object for the aligned sum can be read from /alignment-tool/aligned-frequency-response. If the impulse alignment mode is being used the filtered impulse responses can be read from /alignment-tool/filtered-impulse-response-a and /alignment-tool/filtered-impulse-response-b.

If the alignment command has a "resultUrl" parameter the result of the command will be posted to the URL as a ProcessResult object. If the command does not generate a result to return the ProcessResult will just indicate that the command has completed processing.

EQ defaults

The /eq endpoint provides access to the default EQ settings. The list of equalisers can be read from the /eq/equalisers endpoint, results for a specific manufacturer can be requested by including a manufacturer query value in the url, e.g. ?manufacturer="miniDSP". The list of equaliser manufacturers can be read from the /eq/manufacturers endpoint. The default equaliser to use for new measurements can be read from /eq/default-equaliser and changed by posting to that endpoint. Note that changing the default equaliser will not alter the equaliser selection for any existing measurements.

The default target settings can be read from the /eq/default-target-settings endpoint and changed by a POST or PUT to that endpoint.

The default target level can be read from the /eq/default-target-level endpoint and changed by a POST to that endpoint.

The default room curve settings can be read from the /eq/default-room-curve-settings endpoint and changed by a POST or PUT to that endpoint.

The path to any house curve file being used can be read from the /eq/house-curve endpoint and changed by a POST to that endpoint. It can be cleared by a DELETE at that endpoint or by posting an empty string.

The settings used when matching a response to a target can be read from the /eq/match-target-settings endpoint and changed by a POST or PUT to that endpoint.

A subscription can be added to be notified of progress during EQ matching. That is done by making a POST to the /eq/subscribe endpoint. The body contains a Subscriber object which has a URL string and an optional set of parameters, there are no parameters for EQ subscriptions. For example, to have changes posted to a /progress endpoint a subscription could be:

{
  "url": "http://127.0.0.1:5374/progress"
}

If REW does not get an OK response to an update the subscription will be cancelled. It can also be removed by posting the same Subscriber data to the /eq/unsubscribe endpoint. The list of subscribers can be read from /eq/subscribers. The change notifications are conveyed as a string.

Generator

The /generator endpoint provides full control of REW's signal generator. A GET at the /generator/status endpoint returns its current status, including whether it is enabled, whether it is playing, the current signal selected and the output level. The generator is not enabled if there is no available audio output.

The /generator/signal endpoint is for signal selection and configuration. A GET at the signal endpoint returns the currently selected signal, a PUT selects a new signal. The list of available signals can be retrieved by a GET at /generator/signals.

Each signal has an associated set of configuration options. For the current signal they are at /generator/signal/configuration. The configurations for other signals can be accessed at /generator/signals/{signalname}/configuration. Details of the effect of the configuration settings are in the signal generator help. Note that some settings are shared between signals. For example, the addDither setting for the sine signal is also used by the square signal and the sawtooth signal, the ditherBits setting for the sine signal is also used for other signals that have an option to add dither. Configuration options can be updated collectively or individually by POST or PUT to the configuration endpoint using the setting names and types returned in the GET response. Partial configuration are accepted by both PUT and POST. The full configuration will often contain settings that are specific to a particular mode of a signal and do not need to be included when using other modes, for example the noise configurations have setting for custom noise filtering which are only applicable when using the custom option.

Some signals accept commands, currently these are the tone group of signals which have commands to step to the previous or next one-third octave centre frequency. A GET at /generator/signal/commands returns the list of commands the current signal accepts, which may be empty. The commands for other signals can be accessed at /generator/signals/{signalname}/commands. Only the current signal can accept commands, use POST to send a command to /generator/signal/command.

The generator output level can be read and set at the /generator/level endpoint. The generator supports a variety of units for output level, the list of accepted units is available at /generator/level/units. The default unit is dBFS, if a unit is not specified when setting level dBFS will be assumed.

The generator output frequency can be read and set at the /generator/frequency endpoint. When setting frequency the unit can be Hz, kHz or absent. If absent Hz is assumed. If the current signal does not have a frequency setting a 400 Bad Request status will be returned.

The /generator/invert-second-output endpoint provides access to whether to invert the second output when two outputs are driven. The value can be changed using POST.

The generator has two protection settings that will stop the generator if the input signal exceeds a dB SPL threshold or if excessive clipping is detected on the input. Those settings can be read and configured at the /generator/protection endpoint.

The generator is started and stopped by using POST to send a command to /generator/commands. A GET at the commands endpoint returns the list of commands the generator accepts.

SPL meter

The /spl-meter/ endpoint provides access to the REW SPL meters and the SPL readings. REW can have up to four SPL meters with a Pro upgrade or one without. The meter being addressed is identified by a numeric ID after the endpoint, e.g. /spl-meter/1/ for the first SPL meter. An SPL meter is started or stopped by posting a command to the /spl-meter/{id}/command endpoint. The list of commands can be read from the /spl-meter/commands endpoint.

SPL meters are configured by sending them an SPLMeterConfiguration, which includes a display mode, weighting and filter setting. Note that the display mode only affects the meter display, data read from the meter always include SPL, Leq and SEL values. The list of display modes can be read from the /spl-meter/modes endpoint. The list of weightings can be read from the /spl-meter/weightings endpoint. The list of filters can be read from the /spl-meter/filters endpoint. Here is an example configuration:

{
  "mode": "SPL",
  "weighting": "C",
  "filter": "Slow",
  "highPassActive": false,
  "rollingLeqActive": true,
  "rollingLeqMinutes": 15
}

The last SPL meter update can be read from the /spl-meter/{id}/levels endpoint. That will return the last SPLValues object generated by an update. More commonly levels would be monitored through a subscription. SPLValues includes SPL, Leq and SEL figures.

{
  "meterNumber": 1,
  "weighting": "C",
  "filter": "Slow",
  "spl": 96.9540023803711,
  "leq": 96.91223907470703,
  "isRollingLeq": true,
  "rollingLeqMinutes": 15,
  "leq1m": 96.91223907470703,
  "leq10m": 96.91223907470703,
  "sel": 110.3396232402612,
  "elapsedTime": 22.015999999999995}

A subscription can be added by making a POST to the /spl-meter/{id}/subscribe endpoint. The body contains a Subscriber object which has a URL string. To have SPL updates posted to a /spl endpoint a subscription could be:

{
  "url": "http://127.0.0.1:5374/spl"
}

If REW does not get an OK response to an update the subscription will be cancelled. It can also be removed by posting the same Subscriber data to the /spl-meter/{id}/unsubscribe endpoint. The list of subscribers can be read from /spl-meter/{id}/subscribers.

RTA

The /rta endpoint provides control of REW's RTA. A GET at the /rta/commands endpoint returns the list of commands, to issue a command POST it to the /rta/command endpoint.

The RTA setup can be queried or set via the /rta/configuration, /rta/appearance-configuration and /rta/distortion-configuration endpoints.

The current status of the RTA (whether it is enabled and whether it is running) can be read from the /rta/status endpoint. Rather than polling status with GET a subscription can be added to be notified whenever the RTA status is updated. That is done by making a POST to the /rta/status/subscribe endpoint. The body contains a Subscriber object which has a URL string and an optional set of parameters, there are no parameters for status updates. For example:

{
  "url": "http://127.0.0.1:5374/rta-status"
}

If REW does not get an OK response to an update the subscription will be cancelled. It can also be removed by posting the same Subscriber data to the /rta/status/unsubscribe endpoint. The list of subscribers can be read from /rta/status/subscribers.

The current RTA input level can be read from the /rta/levels endpoint, that returns a list of RTALevel structures. The list has one RTALevel entry unless stereo inputs are being captured, when there are two. The data has a nanosecond timestamp and a running sum of the number of samples processed by the RTA since it was last started or restarted. The default unit for the level data is SPL but other units can be requested by including a unit query value in the url, e.g. ?unit=dBFS. The list of units available can be read from /rta/levels/units. Rather than polling levels with GET a subscription can be added to be notified whenever the RTA levels are updated. That is done by making a POST to the /rta/levels/subscribe endpoint. The body contains a Subscriber object which has a URL string and an optional set of parameters, the parameters can contain the required unit, for example:

{
  "url": "http://127.0.0.1:5374/levels",
  "parameters": {unit: "dBFS"}
}

If REW does not get an OK response to an update the subscription will be cancelled. It can also be removed by posting the same Subscriber data to the /rta/levels/unsubscribe endpoint. The list of subscribers can be read from /rta/levels/subscribers.

The rms and peak RTA responses can be read from the /rta/captured-data and /rta/captured-peak-data endpoints respectively. A FrequencyResponse object is returned, with a nanosecond timestamp, a running sum of the number of samples processed by the RTA since it was last started or restarted, a smoothing setting, a start frequency, a points-per-octave figure (for fractional octave RTA captures) or a frequency step value (for spectrum captures), and a Base64-encoded string of response magnitudes generated from the raw bytes of 32-bit float magnitude values at each frequency. The default unit for the magnitude data is SPL but other units can be requested by including a unit query value in the url, e.g. ?unit=dBFS. The list of units available can be read from /rta/captured-data/units. Rather than polling with GET a subscription can be added to be notified whenever the RTA captured or captured peak data are updated. That is done by making a POST to the /rta/captured-data/subscribe or /rta/captured-peak-data/subscribe endpoint. The body contains a Subscriber object which has a URL string and an optional set of parameters, the parameters can contain the required unit, for example:

{
  "url": "http://127.0.0.1:5374/captured",
  "parameters": {unit: "dBFS"}
}

When multiple inputs are being captured the parameters can contain an input index between 1 and the number of inputs, e.g index: 2. If there is no input index the rms average of the inputs is returned.

If REW does not get an OK response to an update the subscription will be cancelled. It can also be removed by posting the same Subscriber data to the /rta/captured-data/unsubscribe or /rta/captured-peak-data/unsubscribeendpoint. The list of subscribers can be read from /rta/captured-data/subscribers or /rta/captured-peak-data/subscribers.

The current RTA input distortion results can be read from the /rta/distortion endpoint, that returns a list of RTADistortion structures. The list has one RTADistortion entry unless stereo inputs are being captured, when there are two. The data has a nanosecond timestamp and a running sum of the number of samples processed by the RTA since it was last started or restarted. The default unit for the level values is SPL but other units can be requested by including a unit query value in the url, e.g. ?unit=dBFS. The list of units available can be read from /rta/distortion/units. The default unit for the relative distortion values is dB but percent can be requested by including a distortion query value in the url, e.g. ?distortion=percent. The list of relative distortion units available can be read from /rta/distortion/relative-units. Rather than polling with GET a subscription can be added to be notified whenever the RTA distortion results are updated. That is done by making a POST to the /rta/distortion/subscribe endpoint. The body contains a Subscriber object which has a URL string and an optional set of parameters, the parameters can contain the required units, for example:

{
  "url": "http://127.0.0.1:5374/distortion",
  "parameters": {unit: dBFS, distortion: percent}
}

If REW does not get an OK response to an update the subscription will be cancelled. It can also be removed by posting the same Subscriber data to the /rta/distortion/unsubscribe endpoint. The list of subscribers can be read from /rta/distortion/subscribers.

Stepped measurement

The /stepped-measurement endpoint provides control of REW's Stepped measurement features.

A GET at the /stepped-measurement/types endpoint returns the list of measurement types. A GET at the /stepped-measurement/type endpoint returns the currently selected type, to select a type POST it to the /stepped-measurement/type endpoint.

A GET at the /stepped-measurement/frequency-span endpoint returns the current start frequency (Hz), end frequency (Hz) and PPO step for a measurement vs frequency as a SteppedFreqSpan object. To change the span POST a SteppedFreqSpan object to the /stepped-measurement/frequency-span endpoint. The list of accepted PPO values can be read from /stepped-measurement/ppo-values.

A GET at the /stepped-measurement/level-span endpoint returns the current start level (dBFS), end level (dBFS) and level step (dB) for a measurement vs level as a SteppedLevelSpan object. To change the span POST a SteppedLevelSpan object to the /stepped-measurement/level-span endpoint.

A GET at the /stepped-measurement/fft-configuration endpoint returns the current FFT configuration for stepped measurements. To change the configuration POST a SteppedFFTConfiguration object to the /stepped-measurement/fft-configuration endpoint.

A GET at the /stepped-measurement/options endpoint returns the current option settings for stepped measurements. To change the options POST a SteppedOptions object to the /stepped-measurement/options endpoint. Other RTA-related options such as distortion HP or LP should be configured through the /rta endpoint.

A GET at the /stepped-measurement/commands endpoint returns the list of commands, to issue a command POST it to the /stepped-measurement/command endpoint. The Start command requires a settlingTimems parameter and either a leveldBFS, frequencyHz or imdStimulus parameter depending on the type of measurement. Here is an example of starting a THD vs frequency measurement:

{
  "command": "start",
  "parameters": {
    "settlingTimems": 0,
    "leveldBFS": -1.0
  }
}

The progress of a stepped measurement can be read from the /stepped-measurement/progress endpoint. That returns a SteppedProgress object, like the example below

{
  "point": 5,
  "points": 31,
  "message": "5/31, approx 1m 8s remaining",
  "timeRemainingSeconds": 68
}

Rather than polling for progress, a subscription can be added to be notified of progress updates. That is done by making a POST to the /stepped-measurement/progress/subscribe endpoint. The body contains a Subscriber object which has a URL string and an optional set of parameters, there are no parameters for progress subscriptions. For example, to have updates posted to a /stepped-progress endpoint a subscription could be:

{
  "url": "http://127.0.0.1:5374/stepped-progress"
}

If REW does not get an OK response to an update the subscription will be cancelled. It can also be removed by posting the same Subscriber data to the /stepped-measurement/progress/unsubscribe endpoint. The list of subscribers can be read from /stepped-measurement/progress/subscribers.

A subscription can be added to be notified of the distortion results for each point as they are generated. That is done by making a POST to the /stepped-measurement/results/subscribe endpoint. The body contains a Subscriber object which has a URL string and an optional set of parameters. The results are returned as RTADistortion objects, see the RTA section for the associated parameters. there are no parameters for progress subscriptions. For example, to have updates posted to a /stepped-results endpoint a subscription could be:

{
  "url": "http://127.0.0.1:5374/stepped-results",
  "parameters": {unit: dBFS, distortion: percent}
}

If REW does not get an OK response to an update the subscription will be cancelled. It can also be removed by posting the same Subscriber data to the /stepped-measurement/results/unsubscribe endpoint. The list of subscribers can be read from /stepped-measurement/results/subscribers.

Import

The /import endpoint provides access to import functionality.

Importing can take some time, a subscription can be added to be notified of import completion. That is done by making a POST to the /import/subscribe endpoint. The body contains a Subscriber object which has a URL string and an optional set of parameters, there are no parameters for import subscriptions. For example, to have updates posted to a /import endpoint a subscription could be:

{
  "url": "http://127.0.0.1:5374/import"
}

If REW does not get an OK response to an update the subscription will be cancelled. It can also be removed by posting the same Subscriber data to the /import/unsubscribe endpoint. The list of subscribers can be read from /import/subscribers.

Import frequency response

The /import/frequency-response endpoint allows the path to a text frequency response file to be set via a FilePath object. Note that backslashes in file paths must be escaped (meaning a double backslash is required), or forward slashes may be used. FilePath has a string for the path to the response file and an optional string for the channels that should be read from the file, not used for frequency response files. When the file is loaded it will generate a new measurement, that can be monitored by subscribing to the /measurements endpoint. The import may take some time. An import completed message will be posted to subscribers when an import has finished. An internal queue allows multiple imports to be requested without waiting for each to finish, up to 9 imports can be pending.

Import impulse response

The /import/impulse-response endpoint allows the path to an impulse response file to be set via a FilePath object. Note that backslashes in file paths must be escaped (meaning a double backslash is required), or forward slashes may be used. FilePath has a string for the path to the response file and an optional string for the channels that should be read from the file if it has more than one channel. If the channels string is omitted or set to "All" all of the channels will be loaded, otherwise the specified channels will be loaded. Channels are numbered from 1 and channels to be loaded may be specified individually separated by commas or as a range separated by a dash, for example "1, 3, 5" or "2-4" or "1-3, 5, 7". As channels of data are loaded they will generate new measurements, that can be monitored by subscribing to the /measurements endpoint. The import may take some time. An import completed message will be posted to subscribers when an import has finished. An internal queue allows multiple imports to be requested without waiting for each to finish, up to 9 imports can be pending.

Import sweep recordings

The /import/sweep-recordings/stimulus endpoint allows the path to the sweep stimulus file to be read or set. If the file is set successfully the response will contain a summary of the stimulus parameters. If the path has not been set reading the endpoint will return an empty string, otherwise it will return the file path. Note that backslashes in file paths must be escaped (meaning a double backslash is required), or forward slashes may be used.

The /import/sweep-recordings/response endpoint allows the path to the sweep response file to be read or set via a FilePath object. If the path has not been set reading the endpoint will return an empty string, otherwise it will return the last file path set. Note that backslashes in file paths must be escaped (meaning a double backslash is required), or forward slashes may be used. FilePath has a string for the path to the response file and an optional string for the channels that should be read from the file if it has more than one channel. If the channels string is omitted or set to "All" all of the channels will be loaded, otherwise the specified channels will be loaded. Channels are numbered from 1 and channels to be loaded may be specified individually separated by commas or as a range separated by a dash, for example "1, 3, 5" or "2-4" or "1-3, 5, 7". As channels of data are loaded they will generate new measurements, that can be monitored by subscribing to the /measurements endpoint. The import may take some time. An import completed message will be posted to subscribers when an import has finished. An internal queue allows multiple imports to be requested without waiting for each to finish, up to 9 imports can be pending.

Room simulator

The /roomsim endpoint provides full control of REW's room simulator. The /roomsim/room-size endpoint provides access to the room dimensions. The /roomsim/room-is-sealed endpoint provides access to whether the room should be treated as a sealed volume. The /roomsim/absorptions endpoint provides access to the room surface absorptions. The /roomsim/sources endpoint is used to retrieve or set the list of sources in the simulation. The list of recognised source names can be retrieved from the /roomsim/source-names endpoint.

The offsets to mic positions around the main position can be read and set at the /roomsim/mic-posn-offsets endpoint.

The calculation options for the simulator can be read and set at the /roomsim/options endpoint.

Each source can be queried and configured at paths below /roomsim/:src, where :src is one of the recognised source names. The source position in the room can be read or set at /roomsim/:src/position. The configuration of the source (its low frequency extension, enclosure type etc.) can be read or set at /roomsim/:src/configuration.

The frequency response for all sources summed can be read from /roomsim/frequency-response. The response for an individual source can be read from /roomsim/:src/frequency-response. The mic position for which the response is generated must be specified as a query parameter, e.g. ?micposition="Main".