Crates.io | smrec |
lib.rs | smrec |
version | 0.1.4 |
source | src |
created_at | 2023-10-12 16:18:55.465418 |
updated_at | 2023-10-22 17:26:21.708725 |
description | Minimalist multi-track audio recorder which may be controlled via OSC or MIDI. |
homepage | https://github.com/alisomay/smrec |
repository | https://github.com/alisomay/smrec |
max_upload_size | |
id | 1001441 |
size | 119,933 |
Minimalist multi-track audio recorder which may be controlled via OSC or MIDI.
I did this because I needed a simple multi-track audio recorder which I could control via OSC or MIDI.
I didn't want and do not have the resources to use a DAW for this purpose in my setup and wanted mono wave files for each channel organized in a directory per recording by date and time
I'm using this recorder in a setup where I use a Behringer XR18 as an audio interface and a LattePanda 3 Delta as a SBC.
Now let's record some sound! 🔔
cargo install smrec
smrec
uses cpal
as the underlying audio API. cpal
supports WASAPI, DirectSound and ASIO on Windows. However, since cargo
builds binaries from source in the target machine and it is not very straight forward to build cpal
with ASIO support due to asio-sys
build script, there is a pre-build script provided in this repository.
To install smrec
on Windows, please follow these steps in order:
setx /M LIBCLANG_PATH "C:\Program Files\LLVM\bin"
$env:LIBCLANG_PATH
. .\pre-build-win.ps1
this script will download the ASIO SDK, set Visual Studio environment variables and CPAL_ASIO_DIR
variable for the current shell session.cargo install smrec
If you know what you're doing feel free to skip these steps and consult the cpal
documentation.
Pre-built binaries as an alternative are available for Windows here due to the complicated process of building cpal
with ASIO support on Windows currently.
smrec
Runs with the default configuration which is:
ctrl+c
is pressed and the program is interrupted.rec_YYYYMMDD_HHMMSS
in the current working directory and records the audio in that directory.wav
format.chn_XX.wav
where XX
is the channel number.To record for a specific duration, use the --duration
flag and specify the duration in seconds.
The following command records for 10 seconds:
smrec --duration 10
By using the --host
and --device
flag , you can specify the audio host and device to use. The following command uses MacBook Pro Microphone
as the audio device:
smrec --device "MacBook Pro Microphone"
smrec list
By default, all channels of the audio device are recorded. You can specify which channels to include or exclude from the recording by using the --include
and --exclude
flags. These flags can not be used together. The following command records only the first two channels of a 4 channel audio device:
smrec --include 1,2
And this command records all channels except the first channel of a 4 channel audio device:
smrec --exclude 1
as seen in the examples, the channel numbers start from 1 and they can be specified as a comma separated list.
By default, the recording is done in the current working directory. You can specify a directory to record to by using the --directory
flag. The following command records to the ~/Music
directory:
smrec --out ~/Music
smrec
uses the cli arguments for configuration and they precede everything. However, you can configure some aspects (probably more to come) of smrec
by using a configuration file so they replace the default configuration. The configuration file is a toml
file and it is named config.toml
. The configuration file is searched in the following order:
.smrec/config.toml
in the current working directory..smrec/config.toml
in the user home directory.The configuration file can configure:
[channel_names]
1 = "Kick"
2 = "Snare"
3 = "Hi-Hat"
smrec
normally starts recording as soon as it is run. However it also has options for various control methods.
Running, smrec --osc
will not start recording immediately but instead it will wait for an OSC message to start recording.
The default OSC port for receiving and sending is chosen randomly by the os and the default addresses for sending and receiving is 127.0.0.1
and 0.0.0.0
.
After running the command above, the output might look like this:
Will be sending OSC messages to 127.0.0.1:61213
Listening for OSC messages on 0.0.0.0:51014
Currently smrec
does not support IPv6.
In the default configuration:
To configure OSC further arguments could be added to the flag:
smrec --osc "<listen_address>:<listen_port>;<send_address>:<send_port>"
or
smrec --osc "<listen_address>:<listen_port>"
the second form would keep the default send address and port.
smrec --osc "0.0.0.0:18000;255.255.255.255:18001"
will listen for OSC messages on all addresses on port 18000
and send OSC messages to all addresses on port 18001
.
Yes, smrec
can also broadcast OSC messages is the OS and the network allows it.
The messages which smrec
listens for are:
/smrec/start
- Starts the recording, sending a second start will stop the running recording and starts a new one creating a new directory in the specified root./smrec/stop
- Stops the recording if there is a running one.The messages which smrec
sends are:
/smrec/start
- Sent when a new recording is started./smrec/stop
- Sent when a running recording is stopped./smrec/error <string>
- Sent when some errors occur and the error message is transferred a string in the argument.smrec
can also be controlled via MIDI. It can even be controlled via OSC and MIDI simultaneously.
Though smrec
is a simplistic application to serve a single purpose the MIDI communication the options it provides for configuring MIDI is extensive.
Running, smrec --midi
will not start recording immediately but instead it will wait for a MIDI CC message to start recording.
Here is the default configuration:
smrec --midi
is synonymous with smrec --midi "[*[(*,16,17)]]"
which will be explained below.The --midi
flag accepts a string argument which is parsed as a configuration string which configures the input and output.
These strings are separated by a semicolon (;
) and the first part configures the input and the second part configures the output.
Any part could be left out.
Deconstruction:
[..]
is a container for an input or output configuration.[port name[..], ..]
a comma separated list of port names which smrec
will connect to.[port name[(..), ..], ..]
each port name should contain at least one channel/MIDI CC filter configuration.(<channel number>, <cc number for starting the recording>, <cc number for stopping the recording>)
this is the structure of a channel/MIDI CC filter configuration.(1,2,3)
here is an example, this will listen for CC 2 on channel 2 to start the recording and CC 3 on channel 2 to stop the recording. All other messages in that port is ignored. MIDI channels are 0 indexed![my nice port[(1,2,3), ..], ..]
this is how we use that tuple.[my nice port[(1,2,3), (15, 127, 126), ..], ..]
as all the elements we can have multiples of those.[ my first port[(1,2,3), (15, 127, 126), (12,4,5)], my second port[(1,2,3)] ]
here is a valid configuration string. It will listen for CC 2 on channel 2 to start the recording and CC 3 on channel 2 to stop the recording on my first port
and listen for CC 2 on channel 2 to start the recording and CC 3 on channel 2 to stop the recording on my second port
. All other messages in those ports are ignored.Use of '*' and glob patterns:
*
matches any port. Which in the end means all ports.[*[(*,16,17)]]
now should make sense.smrec
can also send midi messages on certain events.
If the output port is configured with a configuration, the configured CC messages will be sent on the configured port and channels on start and stop events.
MIDI CC values are considered momentary.
Once a value 127
is received through a configured MIDI CC number the action is taken immediately.
This is why sending bursts of MIDI CC messages is not a good idea.
Every message would trigger a new recording if it is configured to start the recording.
smrec
sends MIDI CC messages with a value of 127
on start and 127
on stop to the configured MIDI CC numbers if output is configured.
As a last example to get the hang of it, this configuration string will listen for CC 2 on channel 2 to start the recording and CC 3 on channel 2 to stop the recording on my first port
and listen for CC 2 on channel 2 to start the recording and CC 3 on channel 2 to stop the recording on my second port
. All other messages in those ports are ignored. On start and stop events, it will send CC 16 with a value of 127 on channel 2 on my first port
and send CC 17 with a value of 127 on channel 2 on my second port
.
[ my first port[(1,2,3), (15, 127, 126), (12,4,5)], my second port[(1,2,3)] ];[ my first port[(1,2,3), (15, 127, 126), (12,4,5)], my second port[(1,2,3)] ]
I'm going to make sure,
But I don't plan to heavily maintain this project, I'll just make sure that it is usable enough and lives.
x86_64
✅aarch64
✅x86_64
(Should work but not checked yet, will be checked though)aarch64
(Should work but not checked yet, will be checked though)x86_64
✅aarch64
✅Thank you 🙏
It is something I needed to resolve a specific problem and I shared it publicly. I hope it resolves your problem too.