Method 1: Implement through Xunjie Voice Recognition Function online (requires internet connection)#
Note: Limit individual audio files to no more than 20M. Can be refreshed multiple times.
Method 2: Implement through Weizheng Online (requires internet connection)#
Note: 3 chances in total, try to merge the audio into one before uploading.
Method 3: Implement through iDi Cloud Dictation online (requires internet connection)#
Method 4: Log in to Yuelu with your mobile phone number (requires internet connection)#
Method 5: Convert audio to text through converter.app (requires internet connection)#
Method 6: Convert speech to text using Faster-Whisper (no internet connection required)#
- First, download and install FFmpeg suitable for your computer system from Github. Installation tutorial can be found at How to download and install FFmpeg on Windows 10?
- Then, download faster-whisper-GUI.exe from Github and install it as an administrator by right-clicking.
- Next, search and download a model ending with "base" from huggingface, and copy it to a suitable directory folder.
- Run FasterWhisperGUI as an administrator.
- Select "Use local model" and choose the downloaded model file, then click "Load Model".
- If you are using an Nvidia graphics card, select "cuda" in the processing device options.
- Click "Execute Transcription".
- Click the plus sign to select the video file to be transcribed.
- After transcription is complete, click "Jump to whisperX and subtitle editing".
- Click "Save Subtitle File".
- You can also choose the subtitle format to save.