Creating a podcast is still too difficult. One of the main barriers to entry is the editing process. After recording a podcast, the podcast producer needs to line up soundwaves in a digital audio workstation and clip the raw audio files to remove sections that need to be removed. As someone who has edited a lot of podcasts, I know that this is difficult and tedious.

One way of simplifying the editing process is to use speech-to-text to produce a transcription of an audio file, and align the text output with the audio. After that alignment, you have a mapping between the text and the audio–so you can delete text and have the corresponding audio be deleted as well.

SpeechBoard is a project by Craig Cannon and Ramon Recuero Moreno. SpeechBoard is an easy way to edit podcasts by deleting transcribed words that are mapped to an audio interview. In this episode, Craig, Ramon and I discuss how SpeechBoard is built and why this product hasn’t existed until recently. We also discuss the podcast world, which Craig is deeply familiar with as the host of Y-Combinator’s podcast. The YC podcast is one of my favorite shows, and if you like SE Daily, you will probably like the YC podcast, so check it out.

The post SpeechBoard with Craig Cannon and Ramon Recuero Moreno appeared first on Software Engineering Daily.