Which service of aws can be used for recognition of animal voice to text.

Hi,

You can look into using rekognition or sagemaker to try to detect animal audio events and create a transcription of animal sounds.

Rekognition

You can use recognition to identify sounds by converting the audio to a visual format.

The blog uses an example of detecting a fire alarm, you may be able to use as a start to detect animals, this solution will have the limitation of not transcribing the audio.

As rekognition normally uses images, video and text you may need to convert your audio date into a visual representation such as a spectrogram.

The blog goes over: Assembling training and test data, looking for discrete sounds in a spectrogram, End-to-end architecture for sound detection

Detect audio events with Amazon Rekognition - https://aws.amazon.com/blogs/machine-learning/detect-audio-events-with-amazon-rekognition/ What is Amazon Rekognition? - https://docs.aws.amazon.com/rekognition/latest/dg/what-is.html

Sagemaker

You can look into using Sagemaker to create a machine learning model that can be used to detect animals, in addition you could create a model that will process that audio to create a text interpretation of their sound. This second part may run into limits regarding converting animal sounds into a phonetic transcription of the sound.

What Is Amazon SageMaker? - https://docs.aws.amazon.com/sagemaker/latest/dg/whatis.html

Below is an example of audio classification using sagemaker. This example steps through how to set up a sagemaker training job, inference endpoint and customer container images.

https://github.com/aws-samples/amazon-sagemaker-audio-classification-pytorch