- Newest
- Most votes
- Most comments
You can for sure use aws transcribe for such usecases. High level steps are as given below.
1/ Upload your files: You'll first need to upload your MP4 and MP3 files to an Amazon S3 bucket (storage on AWS).
2/ Setup and Start the transcription job: Using the AWS Management Console or SDK, you can initiate a transcription job specifying the uploaded file location, language codes (Norwegian - 'no-NO' and English - 'en-US'), and the media format (MP3 or MP4).
3/ Receive the transcript: Once processing is complete, AWS Transcribe delivers the transcribed text in a JSON format. You can then download the JSON file and use a text editor (like Notepad) to open and print the English text.
This tutorial might come handy for ref flow : https://aws.amazon.com/getting-started/hands-on/create-audio-transcript-transcribe/
You can use the following repository. https://github.com/MichaelShapira/call-center-call-sentiment-tone-summary It transcribes the text from s3. Later it uses Amazon Bedrock to get the sentiment and summary of the audio. But in your use-case you can use only the transcription part
Relevant content
- asked 2 years ago
- asked 3 months ago
- asked a year ago
AWS OFFICIALUpdated 2 years ago- AWS OFFICIALUpdated a year ago
- AWS OFFICIALUpdated a year ago
- AWS OFFICIALUpdated 3 years ago
Thanks, akad.
OK, I have the mp3 file in Amazon S3 bucket and see a screen with details. However, I DO NOT see the TRANSCRIBE option as shown in the hands-on tutorial?! How do I activate TRANSCRIBE option? I finally managed to get transcript of the recording BUT, even though INPUT set to no:NO (Norwegian), output was NORWEGIAN rather than ENGLISH?! Since there was no option to set output language, I assume the default was ENGLISH...but it was NOT! The file to be transcribed has narratives in NORWEGIAN interspersed with songs sung in ENGLISH. Does this type of file content present problems? Thanks.