[FREE] ConvertWithMoss - convert from/to WAV,Bitwig,SFZ,SF2,DecentSampler,MPC/Force,Wave-/Modwave/KMP,NKI,EXS) v9.5.0

[jonlemon]

First and foremost: Thanks so much for your amazing tool! I just discovered it a couple of days ago.

I have encountered a problem when converting Decent Samples or Kontakt libraries to the 1010music format: In some libraries the audio samples start with silence (e.g., https://www.pianobook.co.uk/packs/steinway-grand/) and this shows up in the Decent Samples XML-file information, e.g.,

```
<sample path="Samples/Steinway_A#5_Dyn1_RR1.wav" volume="4.9999995231628dB" tuning="0.00" rootNote="94" start="35806" end="344405" loNote="88" hiNote="94" loVel="1" hiVel="60" />
````

but in the 1010music presets.xml the start point information does not show up.

```
<cell filename=".\Steinway_A#5_Dyn1_RR1.wav" row="0" type="asset">
<params asssrccol="0" asssrcrow="0" keyrangebottom="88" keyrangetop="94" rootnote="94" velrangebottom="1" velrangetop="60"/>
</cell>
```

and as a consequence, there is a noticeable delay between key press and the sample actually sounding, making the preset unplayable.

Is there a way to use this start-point information in the 1010music format? Or is it not supported for multisamples? If the latter is the case, then what's the best way to deal with this? Could your tool drop the silence in the beginning? Or should I, as the user, use some batch tool to cut the silence?

[jonlemon]

Another issue with conversion to 1010music I came across was with libraries that use round robin samples. E.g., again in https://www.pianobook.co.uk/packs/steinway-grand/, there are two samples for each note range and velocity range, used for round robin in the decent sampler. E.g. there's a "Steinway_A#5_Dyn1_RR1.wav" and a "Steinway_A#5_Dyn1_RR2.wav" file that get two entries in 1010music's presets.xml

```
<cell filename=".\Steinway_A#5_Dyn1_RR1.wav" row="0" type="asset">
<params asssrccol="0" asssrcrow="0" keyrangebottom="88" keyrangetop="94" rootnote="94" velrangebottom="1" velrangetop="60"/>
</cell>
```

and

```
<cell filename=".\Steinway_A#5_Dyn1_RR2.wav" row="71" type="asset">
<params asssrccol="0" asssrcrow="0" keyrangebottom="88" keyrangetop="94" rootnote="94" velrangebottom="1" velrangetop="60"/>
</cell>
```

Since they share the same key range and velocity range information, they would be always triggered together, which is not what one would usually want.

1. One solution could be to filter out repeating entries identified based on key range and velocity range. That is, in the above example only "Steinway_A#5_Dyn1_RR1.wav" would be used and exported. "Steinway_A#5_Dyn1_RR2.wav" would not be exported.

2. Another, more ambitious solution would be to use all round-robin samples in the 1010music format to create some variation by partitioning the velocity range. E.g., in the above example the velocity range (1 to 60) could be split into, say, 1 to 30 assigned to "Steinway_A#5_Dyn1_RR1.wav" and 31 to 60 assigned to "Steinway_A#5_Dyn1_RR2.wav". This way, depending on the exact velocity played, one of the two samples will be played, giving slight, additional variation in sound. To really make this more effective, it would be cool to repeatedly subsplit the velocity range, e.g., 1 to 10, 11 to 20 ... 51 to 60 (or even in steps of 5) and assign the samples alternatingly. This way, even very small variations in velocity would trigger different samples.

Both solutions would work also with more than two round-robin samples.

[moss]

First and foremost: Thanks so much for your amazing tool! I just discovered it a couple of days ago.

I have encountered a problem when converting Decent Samples or Kontakt libraries to the 1010music format: In some libraries the audio samples start with silence (e.g., https://www.pianobook.co.uk/packs/steinway-grand/) and this shows up in the Decent Samples XML-file information, e.g.,

...

and as a consequence, there is a noticeable delay between key press and the sample actually sounding, making the preset unplayable.

Is there a way to use this start-point information in the 1010music format? Or is it not supported for multisamples? If the latter is the case, then what's the best way to deal with this? Could your tool drop the silence in the beginning? Or should I, as the user, use some batch tool to cut the silence?

Thanks for reporting! Seems that the start and end were missing (and reverse as well). Will be fixed in the next update!

[moss]

Another issue with conversion to 1010music I came across was with libraries that use round robin samples. E.g., again in https://www.pianobook.co.uk/packs/steinway-grand/, there are two samples for each note range and velocity range, used for round robin in the decent sampler. E.g. there's a "Steinway_A#5_Dyn1_RR1.wav" and a "Steinway_A#5_Dyn1_RR2.wav" file that get two entries in 1010music's presets.xml

```
<cell filename=".\Steinway_A#5_Dyn1_RR1.wav" row="0" type="asset">
<params asssrccol="0" asssrcrow="0" keyrangebottom="88" keyrangetop="94" rootnote="94" velrangebottom="1" velrangetop="60"/>
</cell>
```

and

```
<cell filename=".\Steinway_A#5_Dyn1_RR2.wav" row="71" type="asset">
<params asssrccol="0" asssrcrow="0" keyrangebottom="88" keyrangetop="94" rootnote="94" velrangebottom="1" velrangetop="60"/>
</cell>
```

Since they share the same key range and velocity range information, they would be always triggered together, which is not what one would usually want.

1. One solution could be to filter out repeating entries identified based on key range and velocity range. That is, in the above example only "Steinway_A#5_Dyn1_RR1.wav" would be used and exported. "Steinway_A#5_Dyn1_RR2.wav" would not be exported.

2. Another, more ambitious solution would be to use all round-robin samples in the 1010music format to create some variation by partitioning the velocity range. E.g., in the above example the velocity range (1 to 60) could be split into, say, 1 to 30 assigned to "Steinway_A#5_Dyn1_RR1.wav" and 31 to 60 assigned to "Steinway_A#5_Dyn1_RR2.wav". This way, depending on the exact velocity played, one of the two samples will be played, giving slight, additional variation in sound. To really make this more effective, it would be cool to repeatedly subsplit the velocity range, e.g., 1 to 10, 11 to 20 ... 51 to 60 (or even in steps of 5) and assign the samples alternatingly. This way, even very small variations in velocity would trigger different samples.

Both solutions would work also with more than two round-robin samples.

Option 3: I will talk to the 1010music guys this week at SuperBooth about such things But I am a bit afraid that they think more in the direction of Groove Box than a proper Sampler (for live playing).

[jonlemon]

First and foremost: Thanks so much for your amazing tool! I just discovered it a couple of days ago.

I have encountered a problem when converting Decent Samples or Kontakt libraries to the 1010music format: In some libraries the audio samples start with silence (e.g., https://www.pianobook.co.uk/packs/steinway-grand/) and this shows up in the Decent Samples XML-file information, e.g.,

...

and as a consequence, there is a noticeable delay between key press and the sample actually sounding, making the preset unplayable.

Is there a way to use this start-point information in the 1010music format? Or is it not supported for multisamples? If the latter is the case, then what's the best way to deal with this? Could your tool drop the silence in the beginning? Or should I, as the user, use some batch tool to cut the silence?

Thanks for reporting! Seems that the start and end were missing (and reverse as well). Will be fixed in the next update!

Thanks so much!

[jonlemon]

Another issue with conversion to 1010music I came across was with libraries that use round robin samples. E.g., again in https://www.pianobook.co.uk/packs/steinway-grand/, there are two samples for each note range and velocity range, used for round robin in the decent sampler. E.g. there's a "Steinway_A#5_Dyn1_RR1.wav" and a "Steinway_A#5_Dyn1_RR2.wav" file that get two entries in 1010music's presets.xml

```
<cell filename=".\Steinway_A#5_Dyn1_RR1.wav" row="0" type="asset">
<params asssrccol="0" asssrcrow="0" keyrangebottom="88" keyrangetop="94" rootnote="94" velrangebottom="1" velrangetop="60"/>
</cell>
```

and

```
<cell filename=".\Steinway_A#5_Dyn1_RR2.wav" row="71" type="asset">
<params asssrccol="0" asssrcrow="0" keyrangebottom="88" keyrangetop="94" rootnote="94" velrangebottom="1" velrangetop="60"/>
</cell>
```

Since they share the same key range and velocity range information, they would be always triggered together, which is not what one would usually want.

1. One solution could be to filter out repeating entries identified based on key range and velocity range. That is, in the above example only "Steinway_A#5_Dyn1_RR1.wav" would be used and exported. "Steinway_A#5_Dyn1_RR2.wav" would not be exported.

2. Another, more ambitious solution would be to use all round-robin samples in the 1010music format to create some variation by partitioning the velocity range. E.g., in the above example the velocity range (1 to 60) could be split into, say, 1 to 30 assigned to "Steinway_A#5_Dyn1_RR1.wav" and 31 to 60 assigned to "Steinway_A#5_Dyn1_RR2.wav". This way, depending on the exact velocity played, one of the two samples will be played, giving slight, additional variation in sound. To really make this more effective, it would be cool to repeatedly subsplit the velocity range, e.g., 1 to 10, 11 to 20 ... 51 to 60 (or even in steps of 5) and assign the samples alternatingly. This way, even very small variations in velocity would trigger different samples.

Both solutions would work also with more than two round-robin samples.

Option 3: I will talk to the 1010music guys this week at SuperBooth about such things But I am a bit afraid that they think more in the direction of Groove Box than a proper Sampler (for live playing).

Option 3 is even better!

Yes, I can see that and why 1010music do not think of the Blackbox and the Tangerine as ultraportable multisample players, but somewhat ironically those two products may in fact be the only such devices with that much flexibility. Or would you know other portable products that would support velocity layers and loop points?

I just recently came across a reddit post, which revolved around the same topic: https://www.reddit.com/r/synthesizers/s/DDSgQRSRLw (and here my thoughts on what it means vis-à-vis the Blackbox: https://www.reddit.com/r/synthesizers/s/zKJGRZU3qO).

[moss]

Another issue with conversion to 1010music I came across was with libraries that use round robin samples. E.g., again in https://www.pianobook.co.uk/packs/steinway-grand/, there are two samples for each note range and velocity range, used for round robin in the decent sampler. E.g. there's a "Steinway_A#5_Dyn1_RR1.wav" and a "Steinway_A#5_Dyn1_RR2.wav" file that get two entries in 1010music's presets.xml

```
<cell filename=".\Steinway_A#5_Dyn1_RR1.wav" row="0" type="asset">
<params asssrccol="0" asssrcrow="0" keyrangebottom="88" keyrangetop="94" rootnote="94" velrangebottom="1" velrangetop="60"/>
</cell>
```

and

```
<cell filename=".\Steinway_A#5_Dyn1_RR2.wav" row="71" type="asset">
<params asssrccol="0" asssrcrow="0" keyrangebottom="88" keyrangetop="94" rootnote="94" velrangebottom="1" velrangetop="60"/>
</cell>
```

Since they share the same key range and velocity range information, they would be always triggered together, which is not what one would usually want.

1. One solution could be to filter out repeating entries identified based on key range and velocity range. That is, in the above example only "Steinway_A#5_Dyn1_RR1.wav" would be used and exported. "Steinway_A#5_Dyn1_RR2.wav" would not be exported.

2. Another, more ambitious solution would be to use all round-robin samples in the 1010music format to create some variation by partitioning the velocity range. E.g., in the above example the velocity range (1 to 60) could be split into, say, 1 to 30 assigned to "Steinway_A#5_Dyn1_RR1.wav" and 31 to 60 assigned to "Steinway_A#5_Dyn1_RR2.wav". This way, depending on the exact velocity played, one of the two samples will be played, giving slight, additional variation in sound. To really make this more effective, it would be cool to repeatedly subsplit the velocity range, e.g., 1 to 10, 11 to 20 ... 51 to 60 (or even in steps of 5) and assign the samples alternatingly. This way, even very small variations in velocity would trigger different samples.

Both solutions would work also with more than two round-robin samples.

Option 3: I will talk to the 1010music guys this week at SuperBooth about such things But I am a bit afraid that they think more in the direction of Groove Box than a proper Sampler (for live playing).

Option 3 is even better!

Yes, I can see that and why 1010music do not think of the Blackbox and the Tangerine as ultraportable multisample players, but somewhat ironically those two products may in fact be the only such devices with that much flexibility. Or would you know other portable products that would support velocity layers and loop points?

I just recently came across a reddit post, which revolved around the same topic: https://www.reddit.com/r/synthesizers/s/DDSgQRSRLw (and here my thoughts on what it means vis-à-vis the Blackbox: https://www.reddit.com/r/synthesizers/s/zKJGRZU3qO).

I talked to them and they are totally open to new ideas! Best is if you send your ideas as well to them to show the interest.

[moss]

ConvertWithMoss 9.5 is out!

Get it from https://mossgrabers.de

• Added support to write Soundfont 2.
• All formats
  • Fixed: In rare cases key-ranges could be stored incorrectly if not fully present in the source file
• 1010music format - Writing
  • New: Set samtrigtype to zero if one-shot.
  • Fixed: Writing sample start, length and reverse were missing.
• DecentSampler - Read
  • Fixed: The sub-folder which contains the library/preset was added to the name which could cause issues in the destination format.
• Sf2 - Reading
  • Fixed: Pitch envelope was only set when a filter was present as well.
• TX16Wx - Read
  • Fixed: samples could sometimes not be found on Macos/Linux
• WAV - Read
  • New: Metadata is now read from info sub-chunks and stored in the Comment metadata field.
• WAV - Write
  • Fixed: Update of Broadcast Audio chunk did fail if no date/time metadata was set.
  • Fixed: Destination file name could be empty if 'prefer folder name' was selected.

https://youtu.be/oIG22XGGGnQ

Enjoy!

[Eurikon]

Any chance this tool comes to iPad OS? Wouldbe great and the best converter ever!

[moss]

Any chance this tool comes to iPad OS? Wouldbe great and the best converter ever!

A definite no. Would need to be reimplemented completely.

[topaz]

Any chance of RX2 to SFZ+midi file ?

[audiojunkie]

Thank you for such a useful program!!! It just keeps getting better and better!!!

[metalsvinet]

Great project! I get either silence or white noise when converting to Logic Sampler (EXS). Seems Jürgen is testing EXS in a third party sampler, so I'm just wondering if anyone has it working in Logic?

Sonoma 14.5 / Logic Pro 11

[moss]

Any chance of RX2 to SFZ+midi file ?

Highly unlikely.

[moss]

Great project! I get either silence or white noise when converting to Logic Sampler (EXS). Seems Jürgen is testing EXS in a third party sampler, so I'm just wondering if anyone has it working in Logic?

Sonoma 14.5 / Logic Pro 11

Does it work with none of the output files or can you track it down to certain sample rates, parameters, etc.?
Also, did you try different source formats?