Skip to Content

Changes for page Social Robot

Last modified by Sofia Vlachopanou on 2025/04/24 23:04
From version 4.3
edited by Nikolaos Soumpeniotis
on 2025/03/01 20:34
Change comment: There is no comment for this version
To version 4.2
edited by Nikolaos Soumpeniotis
on 2025/03/01 20:34
Change comment: There is no comment for this version

Summary

Details

Page properties
Content
... ... @@ -12,7 +12,7 @@
12 12  (% style="font-size: 11pt; font-variant: normal; white-space: pre-wrap; font-family: Arial, sans-serif; color: rgb(33, 33, 33); background-color: rgb(255, 255, 255); font-weight: 400; font-style: normal; text-decoration: none" %)Another important technology to consider is Cloud Computing, which enables access to external libraries, enhancing the robot’s ability to interact by enriching dialogues. Additionally, cloud computing allows for real-time monitoring of the robot’s interactions and facilitates adaptive decision-making when needed [1].
13 13  
14 14  (% style="line-height:1.38; margin-top:16px; margin-bottom:16px" %)
15 -(% style="font-size: 11pt; font-variant: normal; white-space: pre-wrap; font-family: Arial, sans-serif; color: rgb(33, 33, 33); background-color: rgb(255, 255, 255); font-weight: 400; font-style: normal; text-decoration: none" %)Equally important aspect of our system is speech recognition, which begins with signal processing through feature extraction. Feature extraction transforms raw audio into a format that machines can process, typically using spectrograms. Spectrograms provide a visual representation of sound frequency over time.
15 +(% style="font-size: 11pt; font-variant: normal; white-space: pre-wrap; font-family: Arial, sans-serif; color: rgb(33, 33, 33); background-color: rgb(255, 255, 255); font-weight: 400; font-style: normal; text-decoration: none" %)Another important aspect of our system is speech recognition, which begins with signal processing through feature extraction. Feature extraction transforms raw audio into a format that machines can process, typically using spectrograms. Spectrograms provide a visual representation of sound frequency over time.
16 16  
17 17  (% style="line-height:1.38; margin-top:16px; margin-bottom:16px" %)
18 18  (% style="font-size: 11pt; font-variant: normal; white-space: pre-wrap; font-family: Arial, sans-serif; color: rgb(33, 33, 33); background-color: rgb(255, 255, 255); font-weight: 400; font-style: normal; text-decoration: none" %)After the feature extraction, Deep Neural Networks (DNNs) play a crucial role in mapping these features to words.Additionally, Convolutional Neural Networks (CNNs) are often utilized to process spectrograms, helping to extract meaningful spatial patterns from the data [3].