Skip to content. | Skip to navigation

Personal tools

Navigation

You are here: Home / Demos / Automatic speech recognition of lectures

Automatic speech recognition of lectures

The following demo is an example output from our ASR system on Jane McGonigal's TED talk on Game Design

ted-demo-automatic-speech-recognition-of-lectures.png

 

The following demo is George Candea's talk on Software performance recorded using the Klewel conference capture system. This one shows final ASR along with n-best output - i.e. the other candidate words for each utterance in the stream.
(see e.g. at ~30s the best utterance is "So it's a loom it so weaving machine that produces coffee" where the correct final word is "cloth" and that appears as an option in the n-best list).

klewel-demo-automatic-speech-recognition-of-lectures.png