Skip to main content

Audio summarize

Audio summarize feature uses AWS Transcribe. It requires the following env variables and IAM permissions.

AWS_REGION​

AWS_REGION=[region]

AWS_ACCESS_KEY​

AWS_ACCESS_KEY=[AWS ACCESS KEY ID]

AWS_SECRET_ACCESS_KEY​

AWS_SECRET_ACCESS_KEY=[AWS SECRET KEY]

AWS_S3_BUCKET_NAME​

S3 bucket is used to save the transcription results of the audio file

AWS_S3_BUCKET_NAME=[bucket name]

IAM permission​

IAM user of the above keys should have the following permission

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "transcribe:*"
            ],
            "Resource": "*"
        },
        {
            "Effect": "Allow",
            "Action": [
                "s3:GetObject"
            ],
            "Resource": [
                "arn:aws:s3:::*transcribe*"
            ]
        }
    ]
}

Additionally, depending on the model selected from the below screenshot, the dependant env variable is required.​

model-list

OpenAI models​

When selecting OpenAI models, please ensure you have the OPEN AI env variables configured. You can find the details in section : OpenAI Setup

Meta & Anthropic models​

Meta & Anthropic models are configured in AWS Bedrock, please ensure you have the AWS env variables configured. You can find the details in section : AWS Setup

Dependant env variable value missing error

When a depandant env variable is not present, you will see an error similar to below. eg: If a OpenAI model was used and OPENAI_API_KEY is missing,

dep-env-error