Tomofon by Klevgrand | Video Giveaway & MPE Demo

Luxthor

@Paulo164 said:

@geremy said:

So I think @Paulo164 that your expectations might be inconsistent with Tomofon’s intent.

Well, let’s say that if I wanted to create a human voice model as Klevgrand did in their factory presets, them I am totally unable to do it with the same pretty good results they achieved. Willing to build similar models as the ones showcased in the factory presets should be at least part of compliant expectations, don’t you think ? I don’t manage to get these results, by far, and hence feel really frustrated…
So what’s left ? Creating weird sounds with unexpected results ? It would be okay if we had after the wavetable stage other synthesis options (FM, ring modulations, etc). But apart from a delay, a reverb and an EQ, that’s all. So I had higher expectations about how good and easy it was to translate my own samples into wavetable models. I still believe the stage where a sample is converted into a wavetable should be improved as this is the key feature of this app.

Did you try to clean and pitch correct voice samples before importing to Tomofon?

Gavinski

Johan from Klevgrand was kind enough to get back to me on this topic:

“ Hard to say why it isn’t working very well without hearing the input audio, but my spontaneous guess is that the voice is out of pitch (wobbling, and/or non-continuous) or maybe the audio quality isn’t good enough - audio sources recorded in a room with lots of reflections with a sensitive and/or omni directional mic would not work very well. Manual import would be the way to force the notes into a certain pitch zone - it’s quite straightforward and described how to do in the manual. There are also videos on our YT-channel showing how to do that. The most important thing is that the audio source is strictly monophonic and has a natural series of overtones - the pitch detection and wavelet extractor works pretty much like a standard autotuner; if Antares Autotune (or Logic’s, or other similar pitch detection based plugins) doesn’t work with the input audio, Tomofon won’t either.. I’m aware one could make improvements on the pitch detection parts (pitch hinting from user, FFT based separation of harmonic and non-harmonic audio components etc), and it might be added in future versions.

Our acoustic instruments and voices in the factory presets batch were recorded in a damped recording studio booth with different directional condenser microphones, and most of the models were created by using automatic or sorted import with some minor edits afterwards.”

geremy

@Gavinski said:
Johan from Klevgrand was kind enough to get back to me on this topic:

“ Hard to say why it isn’t working very well without hearing the input audio, but my spontaneous guess is that the voice is out of pitch (wobbling, and/or non-continuous) or maybe the audio quality isn’t good enough - audio sources recorded in a room with lots of reflections with a sensitive and/or omni directional mic would not work very well. Manual import would be the way to force the notes into a certain pitch zone - it’s quite straightforward and described how to do in the manual. There are also videos on our YT-channel showing how to do that. The most important thing is that the audio source is strictly monophonic and has a natural series of overtones - the pitch detection and wavelet extractor works pretty much like a standard autotuner; if Antares Autotune (or Logic’s, or other similar pitch detection based plugins) doesn’t work with the input audio, Tomofon won’t either.. I’m aware one could make improvements on the pitch detection parts (pitch hinting from user, FFT based separation of harmonic and non-harmonic audio components etc), and it might be added in future versions.

Our acoustic instruments and voices in the factory presets batch were recorded in a damped recording studio booth with different directional condenser microphones, and most of the models were created by using automatic or sorted import with some minor edits afterwards.”

I guess that makes sense in that most of the audio input that I have used has been professionally recorded (lots of BoC!) or directly recorded 'in the box' from other software instruments.

Gavinski

@geremy said:

@Gavinski said:
Johan from Klevgrand was kind enough to get back to me on this topic:

“ Hard to say why it isn’t working very well without hearing the input audio, but my spontaneous guess is that the voice is out of pitch (wobbling, and/or non-continuous) or maybe the audio quality isn’t good enough - audio sources recorded in a room with lots of reflections with a sensitive and/or omni directional mic would not work very well. Manual import would be the way to force the notes into a certain pitch zone - it’s quite straightforward and described how to do in the manual. There are also videos on our YT-channel showing how to do that. The most important thing is that the audio source is strictly monophonic and has a natural series of overtones - the pitch detection and wavelet extractor works pretty much like a standard autotuner; if Antares Autotune (or Logic’s, or other similar pitch detection based plugins) doesn’t work with the input audio, Tomofon won’t either.. I’m aware one could make improvements on the pitch detection parts (pitch hinting from user, FFT based separation of harmonic and non-harmonic audio components etc), and it might be added in future versions.

Our acoustic instruments and voices in the factory presets batch were recorded in a damped recording studio booth with different directional condenser microphones, and most of the models were created by using automatic or sorted import with some minor edits afterwards.”

I guess that makes sense in that most of the audio input that I have used has been professionally recorded (lots of BoC!) or directly recorded 'in the box' from other software instruments.

You mean you actually found a BoC sound without Wow and Flutter lol?

Gavinski

YouTube winner of a free copy of Tomofon: pipster1001

DiscoDSP (Jorge Reales giveaway is still open on YT, Twitter, Insta and Threads btw 🔥. Chances of winning on Threads are ridiculously high lol

Samu

@Gavinski said:
YouTube winner of a free copy of Tomofon: pipster1001

DiscoDSP (Jorge Reales giveaway is still open on YT, Twitter, Insta and Threads btw 🔥. Chances of winning on Threads are ridiculously high lol

Threads is not even available in the EU as they don’t approve suckerbergs app to app integrity violations…

Gavinski

@Samu said:

@Gavinski said:
YouTube winner of a free copy of Tomofon: pipster1001

DiscoDSP (Jorge Reales giveaway is still open on YT, Twitter, Insta and Threads btw 🔥. Chances of winning on Threads are ridiculously high lol

Threads is not even available in the EU as they don’t approve suckerbergs app to app integrity violations…

Yes...fair enough. I'm happy to see the EU cracking down on big tech and data, although I'm not particularly concerned what Facebook does with my data as I am pretty immune to online advertising. I have also seen some posts though stating that the UK being able to access Threads is the only good thing to have come out of Brexit 😂.

Paulo164

@Gavinski said:
Johan from Klevgrand was kind enough to get back to me on this topic:

“ Hard to say why it isn’t working very well without hearing the input audio, but my spontaneous guess is that the voice is out of pitch (wobbling, and/or non-continuous) or maybe the audio quality isn’t good enough - audio sources recorded in a room with lots of reflections with a sensitive and/or omni directional mic would not work very well. Manual import would be the way to force the notes into a certain pitch zone - it’s quite straightforward and described how to do in the manual. There are also videos on our YT-channel showing how to do that. The most important thing is that the audio source is strictly monophonic and has a natural series of overtones - the pitch detection and wavelet extractor works pretty much like a standard autotuner; if Antares Autotune (or Logic’s, or other similar pitch detection based plugins) doesn’t work with the input audio, Tomofon won’t either.. I’m aware one could make improvements on the pitch detection parts (pitch hinting from user, FFT based separation of harmonic and non-harmonic audio components etc), and it might be added in future versions.

Our acoustic instruments and voices in the factory presets batch were recorded in a damped recording studio booth with different directional condenser microphones, and most of the models were created by using automatic or sorted import with some minor edits afterwards.”

This is super insightful, thank you. I would not have been able to get a better reply anyway.
I will check my audio sample against a pitch detection plugin.

Gavinski

Glad it was helpful... Let us know your progress

@Paulo164 said:

@Gavinski said:
Johan from Klevgrand was kind enough to get back to me on this topic:

“ Hard to say why it isn’t working very well without hearing the input audio, but my spontaneous guess is that the voice is out of pitch (wobbling, and/or non-continuous) or maybe the audio quality isn’t good enough - audio sources recorded in a room with lots of reflections with a sensitive and/or omni directional mic would not work very well. Manual import would be the way to force the notes into a certain pitch zone - it’s quite straightforward and described how to do in the manual. There are also videos on our YT-channel showing how to do that. The most important thing is that the audio source is strictly monophonic and has a natural series of overtones - the pitch detection and wavelet extractor works pretty much like a standard autotuner; if Antares Autotune (or Logic’s, or other similar pitch detection based plugins) doesn’t work with the input audio, Tomofon won’t either.. I’m aware one could make improvements on the pitch detection parts (pitch hinting from user, FFT based separation of harmonic and non-harmonic audio components etc), and it might be added in future versions.

Our acoustic instruments and voices in the factory presets batch were recorded in a damped recording studio booth with different directional condenser microphones, and most of the models were created by using automatic or sorted import with some minor edits afterwards.”

This is super insightful, thank you. I would not have been able to get a better reply anyway.
I will check my audio sample against a pitch detection plugin.

Paulo164

@Gavinski said:
Glad it was helpful... Let us know your progress

@Paulo164 said:

@Gavinski said:
Johan from Klevgrand was kind enough to get back to me on this topic:

“ Hard to say why it isn’t working very well without hearing the input audio, but my spontaneous guess is that the voice is out of pitch (wobbling, and/or non-continuous) or maybe the audio quality isn’t good enough - audio sources recorded in a room with lots of reflections with a sensitive and/or omni directional mic would not work very well. Manual import would be the way to force the notes into a certain pitch zone - it’s quite straightforward and described how to do in the manual. There are also videos on our YT-channel showing how to do that. The most important thing is that the audio source is strictly monophonic and has a natural series of overtones - the pitch detection and wavelet extractor works pretty much like a standard autotuner; if Antares Autotune (or Logic’s, or other similar pitch detection based plugins) doesn’t work with the input audio, Tomofon won’t either.. I’m aware one could make improvements on the pitch detection parts (pitch hinting from user, FFT based separation of harmonic and non-harmonic audio components etc), and it might be added in future versions.

Our acoustic instruments and voices in the factory presets batch were recorded in a damped recording studio booth with different directional condenser microphones, and most of the models were created by using automatic or sorted import with some minor edits afterwards.”

This is super insightful, thank you. I would not have been able to get a better reply anyway.
I will check my audio sample against a pitch detection plugin.

Yes, I will. Thanks.