All commands below relies on the latest version of aws-cli.

  • macOS install:

    $ brew install awscli
    $ aws --version
    aws-cli/2.0.19 Python/3.8.3 Darwin/19.5.0 botocore/2.0.0dev23


  • Register default profile:

    $ aws configure
  • Register additional profile:

    $ aws configure --profile bob
  • List access keys:

    $ aws iam list-access-keys
  • List access keys of another profile:

    $ aws iam --profile bob list-access-keys


  • Set media types (formerly known as MIME types) of all atom files found in a feed folder:

    $ aws s3 cp s3://my-bucket s3://my-bucket --exclude '*' --include '*feed/index.atom' --recursive --no-guess-mime-type --content-type "application/atom+xml" --metadata-directive "REPLACE"


  • Fetch all names of the first 100 transcription jobs:

    $ aws transcribe list-transcription-jobs --query '[TranscriptionJobSummaries[*].TranscriptionJobName]' --max-results 100 --output text
  • Get URL of the transcript produced by my_job_name job:

    $ aws transcribe get-transcription-job --transcription-job-name my_job_name --query '[TranscriptionJob.Transcript.TranscriptFileUri]' --output text
  • Same as above but save the transcript content directly to a local transcript.txt file:

    $ AWS_PAGER="" aws transcribe get-transcription-job --transcription-job-name my_job_name --query '[TranscriptionJob.Transcript.TranscriptFileUri]' --output text | wget -i - -O - | jq --raw-output '.results.transcripts[0].transcript' > transcript.txt
  • Putting it all together, here is how do download all transcripts from all your jobs:

    for JOB_ID in $(aws transcribe list-transcription-jobs --query '[TranscriptionJobSummaries[*].TranscriptionJobName]' --max-results 100 --output text);
        do AWS_PAGER="" aws transcribe get-transcription-job --transcription-job-name "$JOB_ID" --query '[TranscriptionJob.Transcript.TranscriptFileUri]' --output text | wget -i - -O - | jq --raw-output '.results.transcripts[0].transcript' > "$JOB_ID".txt;