In this webinar, aired on 17 October 2018, we discussed the key challenges that media companies face when using automatic speech recognition technology, why these challenges still remain and how we as speech technology providers are working to overcome them. Some of the key challenges we covered included:
- Dealing with uncommon vocabulary
- Issues with punctuation
- The use of specific dialects and accents
The speakers were:
Machine Learning Engineer at Speechmatics
Will holds a masters’ degree in computer science and has been working in the speech recognition field for 4 years. His research has focused on developing GPU-based machine learning models and deploying end-to-end speech systems in the cloud.
Emerging Technology Lead in Access Services at Red Bee Media
Hewson Maxwell is a technology expert with over fifteen years’ experience in the Broadcast industry. He has worked for the BBC, Red Bee Media, and Ericsson supporting and developing a variety of broadcast systems, and during the last eight years has specialised in Access Services technology. Hewson’s current role is “Emerging Technology Lead – Access Services” for Red Bee Media, an Ericsson company. The role involves leading an R&D and development team working on the incorporation of ASR, Deep Learning and other cutting-edge technologies into new products and existing production workflows.
Just sit back, relax and enjoy the webinar!