Stefan Lederer of Bitmovin lays out several areas where AI is currenly excelling in saving time and money in the media industry
A lot of the processes in media workflows are labour and time intensive and lend themselves particularly well to being automated.
While deploying AI-enabled tools could cost time and money initially, there is certainly a huge potential for cost savings, as well as for enabling new features or revenue streams.
In a survey we recently conducted, 37% of respondents said they are interested in introducing AI to tag and categorise video.
Audio transcription and speech-to-text is another huge area, with 55% of respondents interested in using AI to make that process more efficient.
At the same time, AI has a huge potential to improve the video experience. Video providers often use analytics tools to track, monitor and analyse their video streams in real-time. This gives them actionable insights into both user behavior and video performance. However, these tools generate huge volumes of data that can be challenging to analyse, especially in real-time.
By using AI to analyse the data being gathered, video providers are able to both enhance video quality and deliver a more engaging experience.
Having the data analysed by AI makes it possible to quickly see, and fix, performance problems, and user behavior analytics can be used to better personalise the experience for individual viewers.
According to our survey, 39% of respondents expect to use AI or ML to improve personalisation and 34% anticipate it will be used to improve recommendations.
AI also has the potential to take the quality of experience even further with super resolution – the process of enhancing the quality or increasing the resolution of a video beyond its original resolution.
AI and ML-based methods involve training deep neural networks with large libraries of low and high-resolution image and video pairs. The networks learn to map the differences between the pairs, and after enough training they are able to accurately generate a high-resolution image from a lower-resolution one.
In the coming months, it is certain we will see an increase in adoption of AI tools across the entire industry. AI can never replace the need for human interaction, but it will certainly help to optimise workflows, analyse data more quickly than could otherwise be possible, and ultimately enable video providers to enhance the video experience for viewers in a cost and time-efficient manner.
Stefan Lederer is CEO and co-founder of Bitmovin
No comments yet