Rev.ai is an accurate automated speech recognition service. You can use Rev.ai as a step to get back a transcript, to get back an SRT subtitle/caption file, or to burn-in subtitles onto your video.

Here's how to use the Rev.ai step.

Assume you want all videos that upload to a specific AWS S3 bucket/folder to be transcribed. The first step will be the S3 trigger step.

1. Then add Rev.AI Transcription step to the Flow

2. Select "Transcribe" as the action

3. Setup the Transcript Options such as entering the correct language code

Source: the input video

Language code: language code associated with the video

Dictionary: array of phrases to help the AI transcribe more accurate (optional). Each phrase should be in string format (accept double quotes only)

Rev.ai Access Token

4. Add "Transfer to Storage" step to save the results/output of the Rev.ai step back to your file hosting service like S3.

Source: the source file which will be uploaded

Host: hosting providers: s3 (we support Amazon S3 for now; more options to be added in the future)

Access Key, Secret Key: the AWS credential that has write access to your S3 bucket

Bucket Name, File Name: the destination where the file will be saved. You can save the file to a specific folder and subfolder by using / (as in the example screenshot above). If you leave the name empty or specify the folder only, a unique file name will be used for the uploaded file.

5. Save, Enable then trigger the flow and check the log for the result.

NOTE:

Valid language codes:

Language

ISO 639 Language Code

Arabic

ar

Bulgarian

bg

Catalan

ca

Croatian

hr

Czech

cs

Danish

da

Dutch

nl

English

en

Farsi (Labs)

fa

Finnish

fi

French

fr

German

de

Greek

el

Hebrew (Labs)

he

Hindi

hi

Hungarian

hu

Indonesian (Labs)

id

Italian

it

Japanese

ja

Korean

ko

Lithuanian

lt

Latvian

lv

Malay

ms

Mandarin

cmn

Norwegian

no

Polish

pl

Portuguese

pt

Romanian

ro

Russian

ru

Slovak

sk

Slovenian

sl

Spanish

es

Swedish

sv

Tamil (Labs)

ta

Telugu (Labs)

te

Turkish

tr

Did this answer your question?