Posts

Amazon Transcribe provides two ways to identify speakers in a piece of audio. In this post you will see how to use channel identification to identify speakers. In a subsequent post I will show how to use speaker identification for cases where you don’t have the speakers on separate channels.

Sometimes a Lambda function can’t deserialize the JSON it receives because the model is wrong or the JSON is not as expected. This post shows how to output the raw JSON to CloudWatch logs so you fix the problem.

Speech to text is now as simple as uploading a file, starting a transcription job, and waiting for the results.

A simple example of uploading a file to S3 with .NET using a few lines of code.

For years I’ve been manually installing the .NET SDK on Linux. There is no automated way to remove them, but deleting everything and starting clean is very easy.

This post shows how to extract key-value pair from fields in a multi-page PDF with Textract.

Textract can do more than just extract plain text, it can also identify fields in a form and extract the key-value pairs. In this post, I’ll show how to use it to extract key-value pairs from a tax document.

Occasionally you need to extract text from an image. Amazon Textract makes this easy. In this post, I’ll show how to use it to extract text from a single image.

Amazon Polly can convert text to speech in a variety of voices. In this post, I’ll use it to read a file containing a dialogue from the play Julius Caesar where Cassius and Brutus discuss Caesar, using a different voice for each speaker.

It’s easy to string together Amazon Translate and Amazon Polly to translate text, and then speak it in a different language.