[LAU] Dereverberation

[Ivica Ico Bukvic]

Thank you, Giso, for sharing this.

Given in my case I am not concerned with mangling the audio (it is not 
intended to be played back, only used to calculate TDOA between mics), I 
have no reservations towards resorting to destructive kinds of 
processing as long as it improves my ability to identify accurate TDOA 
between multiple mics.

Best,

Ico


On 9/12/2016 5:07 PM, Giso Grimm wrote:
> On 09/11/2016 10:14 PM, Ivica Ico Bukvic wrote:
>> Does anyone have access to an easy to implement (as in code)
>> dereverberation algorithm assuming one has access to the room impulse
>> response? I am looking for a way to clean-up recorded audio signal in
>> order to improve signal clarity (e.g. speech). Any info on this topic is
>> most appreciated.
> a common way to achieve dereverberation in speech processing is to use
> spatially separated microphones (e.g., binaural), and use the
> inter-microphone coherence function as an estimator of the gain. There
> are different approaches of calculating the coherence function,
> basically they are all frequency-domain based (but often with short
> analysis blocks/low resolution), and require some temporal smoothing.
> The artefacts are annoying if set to strong filtering, but subtle
> settings may decrease listening effort and are (especially by hearing
> impaired listeners) preferred over unprocessed signals.
>
> Below I added are some references, you will see that the basic algorithm
> was available already in the 70ies... If you need any of the
> publications or some code hints let me know off-list.
>
> Best,
>
> Giso
>
>
> @Article{Allen1977b,
>    Title                    = {Multimicrophone signal-processing
> technique to remove room reverberation from speech signals},
>    Author                   = {Jont B. Allen and D. A. Berkley and J.
> Blauert},
>    Journal                  = jasa,
>    Year                     = {1977},
>
>    Month                    = oct,
>    Number                   = {4},
>    Pages                    = {912-915},
>    Volume                   = {62}
> }
>
> @Article{Kollmeier1993,
>    Title                    = {Binaural Noise-reduction Hearing Aid
> Scheme with R
> eal-time Processing in the Frequency Domain},
>    Author                   = {B. Kollmeier and J. Peissig and V. Hohmann},
>    Journal                  = {Scand Audiol},
>    Year                     = {1993},
>    Pages                    = {28-38},
>    Volume                   = {Suppl. 38}
> }
>
> @Article{Wittkop2003,
>    Title                    = {Strategy-selective noise reduction for
> binaural digital hearing aids},
>    Author                   = {Wittkop, Th. and Hohmann, V.},
>    Journal                  = {Speech Communication},
>    Year                     = {2003},
>    Pages                    = {111-138},
>    Volume                   = {39}
> }
>
> @Article{Grimm2009a,
>    Title                    = {Increase and subjective evaluation of
> feedback stability in hearing aids by a binaural coherence based noise
> reduction scheme},
>    Author                   = {Giso Grimm and Volker Hohmann and Birger
> Kollmeier},
>    Journal                  = {IEEE Transactions on Audio, Speech and
> Language Processing},
>    Year                     = {2009},
>
>    Month                    = {September},
>    Number                   = {7},
>    Pages                    = {1408-1419},
>    Volume                   = {17},
>
>    Doi                      = {10.1109/TASL.2009.2020531}
> }
>
> _______________________________________________
> Linux-audio-user mailing list
> Linux-audio-user at lists.linuxaudio.org
> http://lists.linuxaudio.org/listinfo/linux-audio-user
>