deepgram.com Speech Recognition API Usage
Since v2.92, support for the deepgram.com speech recognition API has been added. This is a foreign AI service that provides a $200 credit upon registration, enough for a while.
- Open the website https://deepgram.com/, register, log in, and enter the console https://console.deepgram.com/
- After logging in, click the large green "Create API Key" in the console.
Clicking will pop up the following image: 
Write a few English letters in the first text box, and then click the bottom "". The SK will then be displayed. Remember to copy it, as shown below:
- Open Menu--Speech Recognition Settings--Deepgram Window
API Key: Enter the key copied in the previous step into the API Key field.
Silence Duration: You can keep the default value of 200 (200ms). If the speech in the video to be recognized is faster, you can reduce it to 150. If it is slower and has more silence, you can increase it to 500 or 800.
- Note: The Deepgram platform does not support Chinese well. Whether using the subtitles returned directly after processing by Deepgram or re-segmenting based on character-level timestamps, punctuation marks are missing, resulting in suboptimal subtitle segmentation. For optimization, the Alibaba Chinese punctuation restoration model is used for re-segmentation. Please select "Chinese Re-segmentation" in the software interface.