These samples were created with WIDI Recognition System Professional in two steps. An audio file was processed and settings were adjusted several times to get better automated transcription result. Then it was edited manually with the TrueTone™ Editor. It took from twenty minutes to two hours to improve each sample.
Here you can play and download source audio file, automated transcription result, manually edited MIDI file and WIDI settings that were used to get automated transcription. If you download the MP3 file and transcribe it yourself with the corresponding WIDI preset, you should get the same (or very close) MIDI to the one in this table.
|Reinhold Gliere Prelude and Scherzo op. 32.
|Anatoly Lyadov Music cabinet. Waltz.
|Georges Bizet Carmen suite No. 2. Habanera.
|Strings & Tube
|Domenico Cimaroza Serenade. Larghetto.
|Guitar & Flute
Sample files were produced by WIDI Recognition System Standard. Recognition settings were adjusted for each sample, but no manual editing was performed.
|Mauro Giuliani Grand Concert
|Heitor Villa-Lobos Prelude No. 5
|George Gershwin Porgy & Bess
|Dave Brubeck There'll Be No Tomorrow
|Mostly Piano & bass
|Anatoly Lyadov Bagatel op. 30.
|Charles Gounod Ave Maria
|Violin & Piano
|Frederic Chopin Etude f-moll, op. 25, no. 2.
Sample files were produced by WIDI Audio To MIDI VST/AU plug-in. These samples were not edited, but recognition settings were adjusted.
|Ronald Weiss, Standard
|Copyright ©2006 Ronald S. Weiss
How to obtain quality scores with WIDI Recognition System? See the article:
The article describes the process of transcribing a particular mp3 step by step. Each step is illustrated with images and Flash demos. Initial mp3 as well as all MIDI and intermediate files are available for download.