Transcribe Video to Text

In today's fast-paced digital age, the ability to transcribe audio and video to text offers immense value in education, business, and research. Converting lectures and meetings into written format enhances accessibility, aids learning, and streamlines data processing. By offering searchable and easily understandable content, this feature speeds up the process of finding and analyzing information. This helps both individuals and organizations to make quicker, well-informed decisions.

What is Transcribe Video?

The model that processes the data uses a special technology called Whisper, developed by OpenAI, to turn spoken words into written text. You can learn more about Whisper here: Whisper - OpenAI.

For this release, Whisper can understand and transcribe speech in English only. It can distinguish between the speakers if there are multiple people in the video. 

Currently, we support MP4, MKV, and MOV video formats for Transcribe Video. 

How to use Transcribe Video

There are some key points that have to be taken into consideration before using the feature: 

• an Internet connection is needed for the feature to work
• the feature won’t work in the background
• only locally stored files can be transcribed for now
• the model is downloaded to the device and operates locally, ensuring user data remains secure without being sent elsewhere
• enabling this feature may increase battery usage
• while the feature works well with spoken content like lectures and podcasts, it may not be suitable for music and noises, potentially leading to inaccurate results
• this feature is not supported on certain devices: iPhone 11 and older iPhone models; iPad (9th, 8th generation) iPad mini (5th generation), iPad Pro 12.9-inch (4th, 3rd generation), iPad Air (3rd generation), iPad Pro 11-inch (2nd, 1st generation)

Since this is an experimental feature, we encourage users to share their feedback. Your input is invaluable as we continue to refine and improve this technology. You can drop us a line at rdsupport@readdle.com.

Currently, we have 4 entry points for this feature: 

• Actions menu — can be entered via the three dots button.
• Extension — can be accessed from other apps via Share.
• Smart Actions – depending on the file type and the action performed, the banner automatically suggests your further actions.
• Video Actions - can be accessed through the video player.

How Transcribe Video works

1. Choose any of the entry points and enable the feature. 
2. The Whisper neurotechnology model will be downloaded to your device.
3. The audio will be extracted from the video.
4. After this, the transcription process begins. 
5. The transcribed file will be saved to the exact location of the original video.