Thank you! speech to text models can be extremely accurate, but it really depends on the audio you feed it. the beauty here with Vocal Slice is that it doesn't need to be perfect because the user is given the tools to fine-tine the selection via the waveform controls and listening to playback. over time i would like to solve the purely automated workflow though, and make the manual step less necessary.