ElevenLabs, known as the strongest AI voice company in the world, has recently launched a speech recognition model called scribe_v1, which supports transcribing audio in 99 languages into text.
Moreover, the free quota is quite generous, supporting the upload of audio or video files up to 1GB in size per instance.
Using it in the video translation software pyVideoTrans This article introduces two ways to use it, online via the web
Using it in the video translation software
Upgrade to version v0.59 https://pvt9.com/downpackage
Go to this page to create an API key: https://elevenlabs.io/app/settings/api-keys
In the video translation software, go to Menu -- TTS Settings -- Elevenlabs.io and fill in the API key you copied, then save it.
Select Elevenlabs.io in the speech recognition channel to use it.
Using it on the web page
- Go to this webpage https://elevenlabs.io/app/speech-to-text. If you don't have an account, please register with your email. No phone verification, no card binding, and no recharge are required.
- After logging in, click Speech to text on the left side, as shown in the figure below.
- After the transcription is complete, click on the displayed name to enter the transcription results page.