A wrapper for transcription results.
A transcription service wrapper that makes it easier to work with transcriptions.
Transcription services provide a cost-effective way to add accessibility to your audio and video, but they are often a challenge to integrate into your system.
Transcriptor aims to make working with transcriptions easier.
Transcriptor looks at transcription objects as a class.
- An object-oriented approach to Transcriptions, Markers, and Speakers
- Nondestructive manipulation of text and references.
Install transcriptor using pip.
pip install transcriptor
Transcriptor currently supports transcription upload and creating a READABLE Transcription Job Object from their output.
Importing Transcriptions from AWS Transcribe
from transcriptor import amazon, AmazonJob job = AmazonJob(filepath='<filename>', bucket='<bucket_name>', ) job.start() # After Some Time job.status() # If it says COMPLETED job.build()
Alternatively, you can load the object via the TranscriptFileUri
AmazonJob.from_uri()) or the JSON object directly (
A loaded Transcription Job from AWS Transcribe will give you access to
(if included), the provided
Alternatives. You also have the original job
object that you can interact with.
Importing from an SubRip Subtitle (SRT)
Text Transcriptions from the Web can be very unique in style, but the most common format is that of at srt file.
You can load an srt file into transcriptor and use that to interact with the individual markers.
from transcriptor import Job amazon.from_srt('FILENAME.srt')
A loaded Transcription Job from srt files will make
These markers are given in order as a list. This gives you the ability to modify a single Marker or iterate your changes across a range or all of the Marker objects.
- Amazon Transcribe (boto3)
Planned Support For
- Google Speech-To-Text
This and much of the work that I do is made possible by those that sponsor me on github.
Sponsors at the $20/month and higher Level
Thank you to them and all of those that continue to support this project!