7 ms·Cohere launches an open source voice model specifically for transcription3 points by newusertoday 6mo ago