With the December update, Podwriter introduces a range of new features and improvements:
Until now, Podwriter was limited to audio sources. From now on, video transcriptions can also be created. This can be done either by direct upload of a video file or through URL import, see below.
Audio and video files from a variety of websites and URLs can now also be imported. Until now, only podcast RSS feeds were supported, but now transcriptions can also be created from YouTube videos, direct links to audio/video files, and websites that embed audio/video files such as media libraries.
The Podwriter Search now allows searching through all transcripts and their descriptions and titles. By activating the "Semantic Search" option, vector search will also look for phrases that are semantically similar. For even more control, a simple query syntax can be used. More information can be found on the search page.
Podwriter now uses the latest Whisper model which offers slightly more accuracy than the previous versions. The model for speaker recognition has also been improved.
There are also a number of UI improvements, including: