From: Patent Tactics, George Brock-Nannestad


Peter Mew undoubtedly presents the only proper method, because the rhythm 
information is destroyed. However, it is culturally imperialistic, thinking 
from the vantage point of using English. This is a language in which several
variants are already available in speech recognition software. Danish also 
has a very good software available, because it was developed as an aid for 
dyslectics. But I would doubt very much whether it would apply to local 
Rwandan witnesses  ("given the small size of many of the communities in 
Rwanda, this is a very real possibility" - here quoting Ishumael out of 
context). I would be surprised but very happy if such software development 
has already been done.

If the witness statements have been transcribed into a text form, perhaps a 
shortcut might be made by using only synthesis software, thereby creating 
something that will sound very strange, but most likely locally 
comprehensible, because our ears are so good at making sense out of sounds.

I am much more worried about the exposure the whole project gets by 
discussing it on an open list. 

Kind regards,

(2nd attempt)


Peter Mew wrote:

> Hi Ishumael 
> This is not something I have done, I postulate it as an avenue for
> investigation
> I would guess the steps to be
> 1. Play the recording into the computer via software that will create a
> file, I'm sure Google will come up with several applications
> 2. Correct those parts of the text that were not interpreted correctly
> 3. have the computer read the text in an appropriate voice, male. female.
> child etc
> 4. record the output of the computer
> 5. replace by editing, those parts of the original recording that you wish
> to be disguised
> As far as I can see there would be no way to reverse this process, as the
> no inflection or accent information in the intermediate text file
> I admit this is a time consuming method but would guarantee anonimity
> cheers
> -pm
> -----Original Message-----
> From: Association for Recorded Sound Discussion List on behalf of Ishumael
> Zinyengere
> Sent: Fri 16/04/2010 14:59
> To: [log in to unmask]
> Subject: Re: [ARSCLIST] Voice Distortion
> Hi Peter
> Thanks but how is that done and which audio or speech editing software
> do you use 
> --- On Thu, 15/4/10, Mew, Peter <[log in to unmask]> wrote:
> From: Mew, Peter <[log in to unmask]>
> Subject: Re: [ARSCLIST] Voice Distortion
> To: [log in to unmask]
> Date: Thursday, 15 April, 2010, 16:57
> How about converting Speech to text and back again, editing in the
> speech you want disguising
> Cheers
> -pm 
> -----Original Message-----
> From: Association for Recorded Sound Discussion List
> [mailto:[log in to unmask]] On Behalf Of Lou Judson
> Sent: Thu 15 Apr 2010 15:31
> To: [log in to unmask]
> Subject: Re: [ARSCLIST] Voice Distortion
> What a serious problem! The best solution that occurs to me might be
> replacing the voice, as in dubbing films into labguages. Is there budget
> to cover that?
> Or possibly converting it to severe low rate MP3 and back to WAV, before
> the final encoding. I have not tried that so not sure. Or some kind of
> exotic voice processing such as Darth Vader?
> <L>
> Lou Judson * Intuitive Audio
> 415-883-2689
> On Apr 15, 2010, at 2:24 AM, Ishumael Zinyengere wrote:
> Using more complex effects, or combinations of effects, is also  
> problematic, because, no matter how complex it gets, it is still  
> possible to reverse engineer and make the voice recognizable.
> We have considered the possibility of using compression to "destroy"  
> the quality to such an extent that the voice of the witness could not  
> be recognizable. The big problem with this is applying it  
> selectively. We only want to distort the voice of the protected  
> witness, not the voices of the lawyers, judges, interpreters etc.  
> After all of the work that is going into producing redacted versions  
> of the audio material, we don't want the end result to sound "bad".  
> Also, using compression is obviously not an option with the redacted  
> WAV files. (Identifying the portions of the recordings in which a  
> protected witness's voice can be heard is a part of the redaction  
> process.)
> One idea that has been suggested is severely limiting the frequency  
> bandwidth. This is similar to the idea of using compression: making  
> bad quality work for us by "throwing away" information so it cannot  
> be retrieved. Sort of an extreme version of a telephone voice. This  
> seems like it could address most of the issues we've identified: it  
> can be applied selectively and it shouldn't be too disturbing to the  
> listener. We are doing some experiments with this option, but would  
> be very curious to hear if anyone has any experience using this type  
> of vocal distortion. Does it sound like it might work? Any other  
> suggestions?
> Any assistance would be greatly appreciated
> - --------------------------------------------------------------------
> Music from EMI 
> This e-mail including any attachments is confidential and may be legally
> privileged. If you have received it in error please advise the sender
> immediately by return email and then delete it from your system. The
> unauthorised use, distribution, copying or alteration of this email is
> strictly forbidden. If you need assistance please contact us on +44 20
> 7000. 
> This email is from a unit or subsidiary of EMI Group Limited. 
> Registered Office: 27 Wrights Lane, London W8 5SW 
> Registered in England No 229231.
> N --------------------------------------------------------------------