Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[larschance]

I am trying to create another speech pack using micmonster and Audacity.

I used micmonster to create the phrases for an actor. Then I converted each phrase from mpeg to wav format and saved it, eventually creating the whole actor file. e.g. Actor 2 with 24 phrases. The lines all sound okay at this stage.

I then type the programme lines over the figure/number mix so that the sim recognises the lines, complete with original mistakes.

Next I insert the new Actor2 into an existing national samples folder, in this case GB, disabling the old Actor2.

When I play the new line up in QMB the Actor2 comes out as white noise not words. The text appears on screen but not the sound of the voice.

I usually ask Marcost who has been very helpful in the past but is possibly working away at present. I asked at Audacity but the AI could not answer my question.

Can anyone help.

[Draken]

When your audio file is ready in Audacity , you actually have to select " export "  ( and not "save "  ) .

[larschance]

Hi Draken.
 
Agreed. I imported the file, converted from mpeg to wav, then exported back to my laptop. I tried varying the sound as advised by a Youtube tutorial but still heard white noise not words.

[Draken]

It is perhaps the " sounds variations " step which makes the issue .
What kind of variations did you apply ?

[larschance]

I first tried 48000 as the default option, then changed that to 44100 as recommended by a Youtube tutorial. Both ended up with white noise.

[Draken]

Does your project looks something like this before you export it ?

[larschance]

Yes Draken. That is the screen.

I open a recorded voice file from my pc and load it into Audacity. After editing it, i.e. splitting the file, I make sure it is in wav format before exporting each phrase. The edit only means splitting a file into different phrases. When I record a voice I tend to get them to say up to three phrases at a time to save their real time and my recording thumb. e.g. 'clear my tail', 'get him off me', and 'bail out' may be in one recording which I split when editing.

I appreciate your interest and help.

With my previous speech packs for SAAF and CW RAF, Marcost did all the heavy work, bless him. I am learning the process and hope to create some more squadrons for our future delight.

[Draken]

Could you try to export the voice file without splitting it ?
If the result is without noise , it could mean that the splitting  creates the problem .

[larschance]

Good thought but I have already tried that.

The process seems straight forward but I may have missed something. I will check the manual again. they always seem to be written as if you already know what to do.

I remember using a small programme Marcost sent me, which converted mpeg to wav. I am wondering if there is more than one type of wav. Maybe micmonster wav is different to other wav formats. Not sure.

[taly01]

According to Creative WaveStudio the audio format of the il2 wingman voice files are .WAV but use MPEG Layer-3 11025kHz 0bit Mono.  Most other sound files are the same but use 22050 or 44100 kHz.

The gun sounds are .WAV but PCM 44100kHz 16bit Mono.

MicMonster can make .wav (if you have paid version) it would save the extra conversion step you are doing?

[larschance]

Thank you taly01. Useful to know.

I did use 44100kHz at first, as recommended in a Youtube tutorial. It also recommended 32 or 64 bit rather than 16 bit so I kept it to 32 bit. I will try 22050 and if necessary 11025bits next. There may be something lost in the conversion from mpeg. Maybe the bit variants affect it. I am still learning all this. At my age it takes a bit longer to sink in.