Analyzes of the survey conducted with regard to our project

By -


We thought of conducting a survey, in order to gather some information for our research based project. For that we prepared three forms in Sinhala, Tamil and English languages, since they are the most widely used languages in Sri Lanka.


We started distributing our forms through social media on 7th July, 2019 and our expectation is to get 500 responses. At the moment, we have succeeded in getting around 350 responses from all the 3 surveys. Here we have asked 8 questions, which are related to our research.


  • Our 1st question is aimed to know the background of the person who fills the form. In order to get the details, we have given some options like undergraduate, graduate, academic  and other Professional. According to the responses we got from all three languages, the most who filled the form are "undergraduates".


  • Our 2nd question is to know the gender. None had not mentioned that they dislike to state their gender. But we gave such an option in case. In English and Sinhala survey forms, most of the responses were from females and in the Tamil language survey, the majority of the responses were by males. But if we consider the overall result, we can say that the survey is mostly filled by females.


  • Our 3rd question is regarding the age group of the audience who have responded to the survey. For that, we divided a considerable range of age into five groups. Below 18 years, 19 years - 24 years, 25 years - 40 years, 41 years - 60 years and above 60 years were the age categories that we provided. According to the received responses, majority who had participated in the survey were around 19-24 years.

  • In our 4th question we asked them for their mother tongue. Our survey was conducted in three languages, but in SriLanka they locally speak Sinhala and Tamil. We created the English language survey form, because there can be people who can't either understand Sinhala or Tamil languages. Under this survey, the Sinhala form was 100% filled by Sinhalese people, and 99% of Tamils filled the Tamil survey form and 80% of Tamils and 20% of Sinhalese had filled the English survey form.
  • Through our 5th question, our intention was to know whether the public have heard any  inappropriate words in an audio content from media. Here the majority of them answered that they have heard inappropriate words and some have answered like may be. It’s clear that in all the surveys, the majority have stated that they have come across objectionable content in audio, which proves that there is a problem as such.

  • To assist our project, we divided the objectionable words into 5 main types. They are curse words, ethnic and racial insults, sexist insults, religious insults and other. Our 6th question is based on knowing what kind of objectionable words are heard mostly by the public. According to this, we found that in Sinhala language most of the people have heard "religious insults", in Tamil language most of the people have heard "Curse words" and in English most of the people have heard "Ethnic and Racial insults". But people have also heard others types of inappropriate content as well. They too had a considerable percentage of responses, but the majority was as mentioned above.









  • In the 7th question we preferred for an open answer like asking them for any solution for this kind of problem. Most of the people answered like mute that contents, ban those audio contents before they get published, make some rules regarding this issue, use beep sounds, create awareness, block those web pages, and ignore them etc.
  • At last we asked them whether they will use a software which will automatically replace offensive content from media for Sinhala and Tamil languages, and majority said yes. So this reveals that our proposed project is worth to be carried on, because there is actually a need in such a system. Most of the people will certainly find it useful.


Our research survey clearly shows that this kind of problem really exist in the society, and people are expecting some kind of new method to prevent objectionable words in common. There are several censorship standards in the entertainment industry to solve this issue, but Sri Lanka has no such defined automatic system which help editors in removing inappropriate  phrases or words, and replacing them. So as a solution we propose a software for automatic audio replacement of inappropriate content for Sri Lankan locale. This system will help users to selectively filter out the objectionable audio content in an audio input, and replace them with predetermined audio inputs like words or sounds. We believe that this research based project will help to preserve the Sri Lankan culture.

Comments

Post a Comment

Popular posts from this blog

The proposed method and technology

By -
Image
We are proposing a method that can easily find and replace the objectionable content in an audio file using an artificial intelligence model. This method will be useful for the media editors, since they can easily find and replace the objectionable content in the audio file with one click. He can input the audio file and select the objectionable content type. After that our system will automatically find and replace the objectionable content with the predefined words or phrases. The replaced words are intended to match with the voice and the modulation of the original input.  When the program starts to run, the user will have the ability to input an audio file(wav or mp3) to be analyzed. Thereafter the audio clip will be analyzed and filtered out the noise using several filters (eg: Low pass, High Pass, etc.). Following that, the user can choose what specific objectionable audio content should be identified and replaced. The program will provide four options namely...
Read more

System Evaluation

By -
Image
After the models were finalized, the hyperparameters of the classifiers and vectorizers were tuned for optimum accuracy using the pipelining method in Scikit-learn. The classification reports of final chosen models for Sinhala, Tamil and English are depicted in TABLE XV. The performance of the finalized binary classification models was evaluated using the Receiver Operating Characteristic (ROC) Curve in Fig 3, Fig 4. and Fig 5, which display the graphical relationships among the metrics shown in TABLE XIV. With the use of ‘Area Under the Curve’ measure, from Fig 1, Fig 2. and Fig 3, it can be observed that the performance of the classifiers is greater than the classifiers with no power. Fig. 1. Receiver Operating Characteristic Curve of the model for Sinhala. Fig. 2. Receiver Operating Characteristic Curve of the model for Tamil.                               Fig. 3.  Receiver Operating Characteristic Curve ...
Read more

Conclusion | Automatic Audio Replacement of Objectionable Content for Sri Lankan Locale

By -
Image
When considering Sri Lanka, the government had to temporarily ban a few social media platforms several times from 2018 to 2019 to control the violence between the communities which was accelerating via hate speech and racist fake news spreading through social media. Yet, there is no proper mechanism to specifically filter and replace objectionable content in audio for Sri Lankan locale. In addressing this issue, we proposed a system for Sri Lankan locale, which automatically detects and replaces objectionable content in audio. In this proposed system, to selectively filter out the potentially objectionable audio content, the input audio is first preprocessed and converted into text format. The presence of racist, cursing, and sexist objectionable content are detected along with their corresponding locations and timestamps through the filtering mechanism. Afterwards, the detected objectionable content is seamlessly replaced with predetermined audio input. The model was tested against a ...
Read more