To analyze a video file from a CCTV footage including the audio and analyzing the footage

What it does

As of now, this project analyzes the Language, Sentiment, Key Phrases, Entities from Speech or Audio file. This project pulls/downloadsMP3 file, places the file in S3 bucket and then Transcribes the content in MP3 file to text. Then using the AWS Comprehend, it analyzes the text/content

How I built it

Built it using AWS ML Services - Transcribe and Comprehend. Using S3 as Storage resource

Challenges I ran into

AWS has a limit of Text length that can be passed to Comprehend

Accomplishments that I'm proud of

Using Two ML Services. Downloading MP3 file in Python

What I learned

AWS ML Services

What's next for Speech Analysis

Integrating a Video instead of MP3 file. Analyzing the Video file and analyzing the Image and Audio

Built With

Share this project: