Update README.md

README.md

@@ -15,7 +15,7 @@ The recording was conducted in the room with variable reverberation, as shown in
 
 The following table shows the specifications of the 6DoF SELD dataset and the conventional SELD dataset.
 
-The dataset is divided into three subsets ("stat.", "3DoF", and "6DoF") according to the state of self-motion: in "stat." the subject remains seated in a chair; in "3DoF" the subject performs head and body rotation movements while standing, and in "6DoF" the subject performs a 0.5 m rotation movement. They walk, change direction, and turn in a circle of 75 m. In "6DoF," subjects walk, change direction, and turn in a circle of 75 m. The signal-to-noise ratio (SNR) was controlled to 6, 10, and 20 dB. The recording room is equipped with a variable reverberation chamber, with reverberation time controlled in three steps ($T_{60}^{500{rm Hz}} = 0.12, 0.30, 0.41$ sec).
+The dataset is divided into three subsets ("stat.", "3DoF", and "6DoF") according to the state of self-motion: in "stat." the subject remains seated in a chair; in "3DoF" the subject performs head and body rotation movements while standing, and in "6DoF" the subject performs a 0.5 m rotation movement. They walk, change direction, and turn in a circle of 75 m. In "6DoF," subjects walk, change direction, and turn in a circle of 75 m. The signal-to-noise ratio (SNR) was controlled to 6, 10, and 20 dB. The recording room is equipped with a variable reverberation chamber, with reverberation time controlled in three steps ($T_{60}^{500Hz} = 0.12, 0.30, 0.41$ sec).
 
 All microphones used for recording were Hosiden KUB4225 with a sampling frequency of 48k and a bit depth of 16 bits; the motion tracking sensor was a combination of HTC Vive Tracker (2018) and HTC SteamVR Base Station 2.0. Sensor signals were recorded at a non-uniform sampling rate of approximately 40 fps and then downsampled to a uniform sampling rate of 20 fps.