![]() ![]() This can be handy if you want to dictate a few paragraphs of text at once, but want more reliable results than Apple’s built-in speech-to-text feature. Click Stop Recording whenever you’re done. It’ll go with whatever you have set in System Settings > Sound > Input. Unfortunately, MacWhisper doesn’t let you pick from a list of microphones. Make a live recordingĬlick New Recording to start right away. This is really handy if you want to save a transcript for a podcast, video or keynote presentation. Running the Small English model on my Mac mini with M2 Pro, it processed a 70-minute uncompressed AIF file from Logic in just 4 minutes 24 seconds - that’s working at 17.2× speed. MacWhisper supports MP3, WAV, M4A, MP4 and MOV files. Just drag an audio file into the window to start transcribing. ![]() Make a transcription from a video or sound file Drag a file in or paste a YouTube URL. But if you want something you can confidently use without making any edits, MacWhisper Pro unlocks the Large model for €16 (that’s $17.45 at the time of writing). I have maybe one edit to make per minute of talking. I think calling it “good” sells it short - it makes excellent transcriptions. The multilingual model will be more versatile but a bit less accurate. The best one you can use for free is the Small English language model (described as “normal speed with good accuracy”). When you upgrade to Pro, you are given the choice of a few different language models to download with varying degrees of accuracy, The free ones are pretty great as is. The free version is pretty good, but to get the most accurate transcriptions, you should upgrade to the Pro version. The macOS app is a free download, but has limits. MacWhisper is based on OpenAI’s state-of-the-art transcription technology called Whisper, which is claimed to have human-level speech recognition. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |