The role of moderation in voice chat apps — How Wafa app addresses issue of inappropriate content

Moderating an app's chat functionality is the key to ensuring that users can enjoy themselves and feel comfortable while engaging with the app and making connections with others. By implementing effective moderation policies and procedures, organizations can help create a safer and more enjoyable experience for users of voice content. Voice content moderation refers to the process of reviewing and filtering voice content to ensure that it meets certain standards and guidelines.

This is a crucial task in today's world, where the amount of voice content being produced and shared online is growing at a rapid pace.

Given the diversity of voice content, the process of moderating it can be complex and challenging. One of the key challenges in voice content moderation is the ability to accurately understand and interpret the content.

Another challenge is the ability to identify and remove inappropriate or offensive content. Voice content can contain explicit language, hate speech, and other forms of offensive material, and it is the responsibility of the moderators to identify and remove such content. Overall, voice content moderation is a crucial task that helps ensure that the voice content shared online is appropriate and meets certain standards.

To address these challenges, many companies and organizations that produce or host voice content have implemented strict moderation policies and procedures.

One such company, Bengaluru, Karnataka-based LVE Innovations Pvt Ltd., has put a great stress on moderation in its voice-centric Wafa app. Wafa is one of India's early-launched voice-centric apps that is available in diverse local languages. It has already crossed 5 million downloads by users from all walks of life.

This makes voice moderation in the app all the more important. Hence, the team behind the Wafa app takes voice moderation very seriously. But, it is a costly process because computers are not yet able to effectively filter speech.

Towards that end, human intervention is needed to judge the context of conversations. This is taken by humans because AI can function only limited and can't deliver fulfilling output, especially for vernacular languages. Human intervention happens on a shift basis where human AI takes a position in moderating content. There are hundreds of content monitors on a shift basis to listen to on all channels in real-time and are carefully monitored.

Moderators are again of two divisions including the senior moderating team that takes utmost care of almost all the top trending voice rooms. The junior moderators travel in and out of all rooms. This is to ensure Wafa's most important policy of no PPR (No Porn, No Politics, No Religion) is complied with well.

Wafa has user-generated content (UGC) where the users come, have fun, and leave. If the users spot anything against The Community guidelines, they can voluntarily submit a complaint or report to our help room or e-mail us. These users can expect a response from instant to within 24hrs.

Yet another feature of Wafa is that moderators have the full right to freeze if any minor issues are spotted. And if the user repeatedly commits the mistake, his/her account will be banned from Wafa's platform or taken down. In this way, moderators at Wafa act as a cross-functional team that works together on a shift basis to determine areas of improvement and ways to optimize them for maximum value.