Computer King Ada’s team once introduced MacWhisper, a free Mac tool developed based on OpenAI Whisper – it not only supports a variety of file formats, including video formats, but even YouTube videos are a good tool. Recently, it has been further evolved and moved towards version 3.0. Continue reading The free speech-to-text tool MacWhisper has been updated to include a long-form interface and batch functionality.

▲Photo source for this article: Jordi Bruin
Free speech-to-text tool MacWhisper updated to add long text interface and batch function

MacWhisper is a free tool developed by Jordi Bruin. It is not available on the App Store and needs to be obtained from the developer page. In addition to the full functionality of the Pro paid version, including faster models and YouTube direct recognition, you can also choose to use Tiny, Base and Small language models for free recognition – please refer to the detailed download and use instructions.Our previous introduction。

This time it has evolved to MacWhisper 3.0. For previous users, they can choose to update when opening the application. After the update, not only the interface has become more intuitive and easy to use, but also a long text display mode like reading an article has been added – which can quickly switch between three display modes including timecode (subtitles). It also brings the ability to batch output speech-to-text files and DOTE file format (Pro version required).

In addition, the Transcribe Podcast function, which extends from the speaker function and can be more suitable for broadcast usage scenarios, can also be tested in beta in this version – the advanced Pro function that can record directly in the app, including Zoom or FaceTime audio, is not mentioned here. If you like it after using it, you can consider paying to unlock it.

MacWhisper 3.0 reportedly performs 40% faster. However, in addition to the previously mentioned recommendations for use on M1 or above Mac devices. The official update information for this 3.0 update mentioned that it will only support macOS 13.0 (Ventura) and above – the developer said that since he has upgraded to the macOS Sonoma developer beta version, it is said that the difficulty of supporting older versions has increased significantly.
But it also said that if you want to use MacWhisper on an older version of the system, you can download the old v2.21 version and continue to use it.

That is, if you have in-depth use needs for this type of voice-to-text service. In fact, you can also refer to the online tool Vocol AI provided by the Taiwan development team – you can refer to our actual measurementAutomatic verbatim draft + GPT summary key points! Vocol.ai’s most powerful AI conference organization tool uses and shares tutorials. Because they also use the ChatGPT language model to provide similar functions, and also add online collaboration and automatic summary functions, the functions are more comprehensive, and there are also a lot of free points to use, so you can also take a look.
Further reading:
Meta Voicebox AI only needs 2 seconds of sound samples to generate multi-national voices and can speak in different styles

Source: KOCPC Chinese