MiniSedric is a minimal backend service designed to analyze human interactions from audio files using AWS services. The entire infrastructure is set up using AWS CDK.

A FastAPI based chatbot server using OpenAI to respond to textual and audio queries.

Audio (.mp3) -> AI Generated Images (work in progress)

speech to text benchmark framework

A simple UI for Amazon Transcribe

AWS Transcribe evaluation pipeline: bulk-process audio files and view the results

We will create a Lambda function that will trigger whenever an object is placed into an S3 bucket. Specifically, we will trigger a job with Amazon Transcribe speech from an audio file placed into S3.

Demo of live transcription of mic audio in Python with AWS

The program will transcribe the voice to text, translate the text to the target language, and then convert the text to speech.

[Backend] 딥러닝 기반 일치율 수식을 이용한 유튜브 콘텐츠 필터링

A Python script to convert a video to text

Sample Implementation of AWS Transcribe and other related services

Example python scripts to evaluate various ASR methods

Converting story oriented youtube video to transcription.

Think-aloud data processing with Amazon Web Services

AWS Lambda script to transcribe an audio file on upload to an S3 bucket.

AWS Lambda Function which creates a transcribe job, that reads mp3 file and converts it into text format in a json file.

Project developed for the Cloud Computing 2020.1 class. We built a service to add translated subtitles to videos using AWS services.