Skip to content

Free DJ Software Tools

This section groups the DJ software tools written by me, all freely available.

What software tools did you built for Traktor?

This folder contains my Traktor tools and CUE tools. Below is a summary; see here for more details

See also the CMDR editor changes.

  • Traktor_clone_cues
    • Clones cues between physically duplicated files.
    • Is also able to merge traktor NML files.
  • rekordbox_add_beatmarkers
  • CUE_tools
    • Tools to generate CUE files and timestamped tracklists
    • Tools to search a whole set in youtube tabs and to scrape lyrics
    • My mp3tag actions scripts
    • Adaptor scripts to run DJCU and Rekordbuddy in Windows
  • 26ms offsets
    • Research about the 26ms mp3 cue shifts in DJ conversion apps. More info
  • BOME tools:

What features did you add to the CMDR TSI editor?

The CMDR editor did not get new features for 2.5 years. In Jan 2020 I've revived this project.

  • Download: https://github.com/cmdr-editor/cmdr/releases/latest/download/cmdr_tsi_editor_latest.zip
  • Documentation: https://github.com/cmdr-editor/cmdr#2020-improvements

Highlights:

  • TP3 and S4-MK3 support
  • Grid quick filter
  • many new Shortcuts (channel change, modifier rotation
  • FX list and encoder mode fixes; same sorting as controller manager

CMDR ChangeLog

How can I run CMDR in my macOS?

There is three ways to edit your mappings in macOS.

Simplest: Xtrememapping

The simplest answer is: buy a copy of XtremmeMapping for macOS.

More complex: Azure Cloud

The more complex answer is: get a free trial of a windows virtual machine.

This will run in the "azure" microsoft cloud, and requires giving your credit card to detail to microsoft (they say over and over that no automatic billing will ever happen after the trial period without you changing your )

installation:

  • link1: https://microsoft.github.io/AzureTipsAndTricks/blog/tip246.html
  • link2: https://www.techrepublic.com/article/build-your-own-vm-in-the-cloud-with-microsoft-azure/

Then connect to your cloud machine using RDP for mac: https://www.techrepublic.com/article/pro-tip-remote-desktop-on-mac-what-you-need-to-know/

Then:

* Install CMDR as explained here: [CMDR installation instructions](https://github.com/cmdr-editor/cmdr#download-and-installation)
* Copy your TSI into the virtual machine (simplest is to use eg google drive on the browser)

Most complex: VirtualBox

The most complex answer is: get a personal virtual machine couples with an evaluation copy of any Windows OS.

Step by step instructions are on: * https://towardsdatascience.com/how-to-install-a-free-windows-virtual-machine-on-your-mac-bf7cbc05888e * Step 1: Download VirtualBox * Step 2: Grab Windows 10 * Step 3: Install VirtualBox and the extension pack * Step 4: Get your OS up and running

Then:

* Install CMDR as explained here: [CMDR installation instructions](https://github.com/cmdr-editor/cmdr#download-and-installation)
* Copy your TSI into the virtual machine (simplest is to use eg google drive on the browser)

original blog post

How to run macOS on windows and linux

I've successfully ran MacOS in a windows machine usig virtualBox:

If you are in linux you can get a docker image premade:

Future windows WSL2 will be able to run docker images as well. Requirement is windows 2004 (from april 2020)

How I build perfect tracklists using CUE files

Update 2022:

  • source of truth: TL.TXT
  • for the first time, create a file tl.tracklist with a header
  • call cue_merge_cues BASE tl.tracklist -n -f
  • hear in foobar
  • check wrong names, fix in tl
  • repeat
  • source of truth: CUE FILE
  • call cue_merge BASE (without TL)
  • add ID3 header in winamp

I use a set of tools to generate a CUE file with the timings of my sets. Once I have this file, I can generate tracklists with timestamps like in this example: https://www.mixcloud.com/dj_estrela/mix-17-cd07-trance-jun-2019/

Steps BEFORE the set (for prepared sets):

  • group the files in folders, per style (Vocal Trance, Uplifting trance, etc)
  • select the tracks and their order using winamp; Once this is OK, run "cue_renumber_files.py" and "cue_make_tracklist.sh".

Steps AFTER the set (both live sets and prepared sets):

  • convert the NML to a text tracklist using https://slipmat.io/playlists/
  • listen the recording in winamp to spot major mistakes. Tag the locations in MMM:SS format. At the end, use "cue_convert_timestamps.sh" to convert to HH:MM:SS format
  • open the huge WAV file in Adobe audition, and perform the following:
    • normalize volume of all tracks
    • fix any obvious mistakes if necessary (eg, track ended too early when playing live, etc)
    • tag the divisions of the tracks inside the wav file
  • convert the tags inside the WAV into a CUE file, using this software: http://www.stefanbion.de/cuelisttool/index_e.htm
    • note: this software fails on files bigger than 2Gb (https://forums.adobe.com/thread/309254). Workaround is splitting the file at the 3hour mark, exactly, then use an option in cue_merge_cues.py to add this offset back in the second file
  • convert the tags inside the WAV into a CUE file, using this software: http://www.stefanbion.de/cuelisttool/index_e.htm
  • merge the CUE file with the Tracklist file using cue_merge_cues.py
  • upload the mix to http://mixcloud.com/dj_estrela
  • transfer the mix automatically to https://hearthis.at/djestrela/#sets

Overview of the Cue tools:

  • cue_renumber_files.py:
    • renumbers mp3 files, in sequence. This is useful to make a sequenced playlist in your operating system folders, outside Traktor.
  • cue_make_tracklist.sh:
    • from a folder, generates basic tracklist text files
  • cue_convert_timestamps.sh:
    • convert MMM:SS to HH:MM:SS format. Winamp uses the first format, Adobe audition uses the second
  • cue_merge_cues.py:
    • this is the main tool. It merge back and forth any CUE file with any tracklist text file. It can read either case from separate or single files. It also cleans the artist - title fields, and generates timestamped tracklists
  • cue_rename_cue.sh:
    • matches the CUE file contents with the FILE tag. This is useful when you rename the files externally.

How much time do you take to make each mix

A LOT of time. Much much more than simply the running time of each mix.
I'm always easily preparing/recording/finishing say 8 mixes at a time, so the whole process for any given mix can easily take months.

Why so much time? Is is worth it?
You tell me :). Check for yourself in https://djestrela.com/. List of styles are in the bottom of this post.

In short all my mixes are carefully planned to tell a story, by focusing on a specific overall theme or sub-sub-style. All my mixes are post-mastered with a single loundness, and feature a clean tracklist and whole lyrics.

I've built some software tools to speedup some parts of this process.

Major styles in https://djestrela.com/:

  • Trance
    • Vocal Trance
    • Uplifitng Trance
    • Tech Trance
  • 80s:
    • Italo Disco
    • Rock FM
    • "regular" 80s Pop and Rock
  • Top-40
    • Commercial House, per decade
    • Latin / Reggaeton

See below for a much more detailed version of this post.

How much time do you take to make each mix - Worst case

In the worst case I take months to do a single 90 minute set. This post describe all steps of such mix.

Preparation:

  • find music:
  • (Tb)
  • collect
  • group the files in folders, per style (Vocal Trance, Uplifting trance, etc)
  • select the tracks and their order using winamp; Once this is OK, run "cue_renumber_files.py" and "cue_make_tracklist.sh".

TODO: finish this post, add picture

What Programming libraries and technical scripts did you author?

Please see my open-source github repos in https://github.com/pestrela?tab=repositories See also my windows shortcuts

Summary:

Open-Source useful information

Below some links useful to build open source projects

Regarding DJ software, it useful for companies to jumpstart their innovation with open source technology. For new stuff like STEMS realtime generation it is far quicker to use this, than to developing your own technology.

The MIT license choose by Spleeter its very permissive and has very few conditions: https://choosealicense.com/licenses/

This is a general example where you just need to attribute things:\ "Third Party Software Acknowledgements" https://www.denondj.com/product-legal-sc5000-prime

Compare this to the GPL, where you need to make your own code that extends it GPL as well: https://www.pioneerdj.com/en/support/open-source-code-distribution/gpl-lgpl-source-code/

Github Markdown tools

This is a small program to check relative links inside the same MD document: md_check_relative_links.py.

What shortcuts you added for Youtube, Google and Discogs?

This folder contains Windows tweaks and other generic tools.

My autohotkey file (AHK) has all kinds of useful global shortcuts that work for any text of any program.\

All these shortcuts operate in a single line, or a list of lines:

  • CTRL+F08: Search in Google
  • CTRL+F09: Search in File Explorer
  • CTRL+F10: Search in Soulseek
  • CTRL+F11: Get current URL
  • CTRL+F12: Search in Youtube (first hit)
  • Win+backspace: toggle any window to be "always on top"
  • Win+Arrows: Arrange windows on the corners, 3 possible sizes. Also supports regular half-screen
  • Win+Insert: Append a copy to the clipboard. Really useful to collect things around and paste once

Using Shift you get varitions of the above:

  • CTRL+SHIFT+F08: Search in Discogs
  • CTRL+SHIFT+F11: Get all tabs url
  • CTRL+SHIFT+F12: Search in Youtube (list)

More scripts in this folder:

  • a youtube-dl wrapper
  • a tool to search 1001tracklists, ticketswap and DJ TechTools maps.

21st century software issues

The 21th century has very clear trends where software cost go down every year. The main factor of this is simplification and lower quality / minimum viable product, many times at the user burden. Some examples follow.

Required Updates:

To reduce cost, both backwards and forward compatibility are dropped. Its far far simpler to the developer to just force the user to upgrade all components if just one them gets upgraded. Instead of checking what actually changed, and if the old/unchanged functions would just work fine with the previous version.\
Typically the server gets updated first, then the client App needs to follow.

Some years ago changes and all protocols had compatibility in mind, so that the user could continue using the previous version as long as possible (except urgent security updates).

Poor Documentation:

Today apps are far far simpler to use for the simple tasks. That is great.

However, the complex tasks are being dumbed down by force, and worse, they are no longer documented structurally anymore.

The existing documentation all moved online and is searchable. That is reasonable, but typically there is no longer structured manuals that would explain the features of eg your new mobile phone, and every single configuration option.

Other examples are the dumbing down of laptops to approach tablets by force (Windows 8 comes to mind). A Tablet is great for content consuming, but a laptop is irreplaceable for content producing.\
Specific examples are Function and insert keys being secondary, the loss of physical buttons like ESC in mac, short keyboards without insert/delete section like it was present on the standard 102 keyboard

Subscriptions / Ownership:

Every year there is fewer laptop users. The "buy once, use forever" model is no longer profitable to small and medium businesses that make unique software. The only solution was to ask money for newer major versions of the software. Every year this gets more and more difficult.

The alternative is subscriptions that stop working if you stop paying. Another worrying trend is companies lawyers that removed the notion of ownership, giving you an authorization instead.

Yet another worrying trend is hardware features that are unlocked by software. Back in the days, the price segmentation was clearly made because of hardware limitations: it was physically impossible to send more data / driver faster / etc. Today, the hardware has gained massive leaps in specifications, but then they come lobotomized by default, and sold as a software upgrade.

Software Quality

on the topic of software quality, please see this blog post:

https://pestrela.github.io/dj_kb/free_dj_tools/#21st-century-software-issues

Topics:

  • Required Updates:
  • Poor Documentation:
  • Subscriptions / Ownership:

Knowledge base organization

I'm looking for a solution to organize this knowledge base :)

Any help is welcome.

Requirements:

  • Have everything in a single markdown page
  • TOC
    • Left sidebar always visible
    • auto generated
    • Expandable sections
  • All headers with link anchors
  • "git push" triggers pipeline

Non-requirements:

  • chronological blog posts
  • separate files
  • categories
  • multiple users
  • user comments

Candidates: