Processing From S3

Audo's noise removal endpoint supports processing audio directly from S3 through the use of presigned URLs. To do this, simply pass presigned URLs to the input and output parameters of the /noise-removal endpoint. The specific format of these presigned urls is as follows:

• input: A presigned GET URL
• output: A presigned PUT URL with correct ContentType (ie. video/mp4)

Note: As of writing, the aws presign command does not support generating PUT URLs. However, you can see the example below for how to generate them using Python.

Full Python Example

First, make sure to have dependencies installed (make sure to upgrade audoai-noise-removal):

pip3 install --upgrade audoai-noise-removal boto3

# Configure aws with `aws configure`

Then, we can use this example script to generate the presigned URLs and remove noise:

bucket = 'my-bucket'

s3_input_file = 'my-input.webm'

s3_output_file = 'my-noise-removed-output.mp4'

output_content_type = 'video/mp4'

audo_api_key = 'MY-AUDO-API-KEY'

# Generate Presigned URLs

import boto3

s3_client = boto3.client('s3')

input_url = s3_client.generate_presigned_url(

'get_object',

Params={'Bucket': bucket, 'Key': s3_input_file}

)

output_url = s3_client.generate_presigned_url(

'put_object',

Params={

'Bucket': bucket,

'Key': s3_output_file,

'ContentType': output_content_type

},

HttpMethod='PUT'

)

# Remove Noise

from time import sleep

from audoai.noise_removal import NoiseRemovalClient

noise_removal = NoiseRemovalClient(api_key=audo_api_key)

job_id = noise_removal.create_job(input=input_url, output=output_url)

noise_removal.wait_for_job_id(job_id)

print('Noise removal complete. Waiting for upload...')

sleep(2.0)

print('Final result:', noise_removal.get_status(job_id))