Skip to main content

Table 1 Overview of processing strategies employed by submitted systems

From: A summary of the REVERB challenge: state-of-the-art and remaining challenges in reverberant speech processing research

 

Task

Characteristics of enhancement part

Characteristics of recognition part

 

SE

ASR

# of channels

Derev.

Advanced

NN-based

Feat./model

Advanced

   

1-ch

2-ch

8-ch

 

feature

AM

adaption

decoding

Alam [17]

 

x

x

  

x

x

x

 

x

Astudillo [18]

 

x

  

x

x

  

x

 

Cauchi [19]

x

x

x

 

x

x

    

Delcroix [20]

x

x

x

x

x

x

 

x

x

x

Epain [21]

x

   

x

x

    

Feng [22]

 

x

x

 

x

x

x

 

x

 

Geiger [23]

 

x

  

x

x

 

x

x

x

Gonzalez [24]

x

 

x

  

x

    

Hirsch [25]

 

x

x

  

x

    

Kallasjoki [26]

 

x

x

 

x

x

  

x

 

Kondo [27]

x

 

x

  

x

    

Leng [28]

x

x

x

 

x

x

 

x

x

 

Lopez [29]

x

 

x

  

x

    

Mimura [30]

 

x

x

  

x

 

x

  

Mitra [31]

 

x

x

   

x

 

x

x

Moshirynia [32]

x

 

x

  

x

    

Ohtani [33]

x

 

x

  

x

    

Palomaki [34]

 

x

x

  

x

  

x

 

Parada [35]

 

x

x

   

x

 

x

 

Tachioka [36]

 

x

x

 

x

x

 

x

x

x

Veras [37]

x

 

x

  

x

    

Wang [38]

x

x

 

x

 

x

  

x

 

Weninger [39]

 

x

x

 

x

x

 

x

x

x

Wisdom [40]

x

 

x

x

x

x

    

Xiao [41]

x

x

x

x

x

x

 

x

x

 

Xiong [42]

 

x

x

  

x

  

x

x

Yu [43]

x

   

x

x