Deepgram Python Captions

This package is the Python implementation of Deepgram's WebVTT and SRT formatting. Given a transcription, this package can return a valid string to store as WebVTT or SRT caption files.

The package is not dependent on Deepgram, but it is expected that you will provide a JSON response from a transcription request from either Deepgram or one of the other supported speech-to-text APIs.

Installation

pip install deepgram-captions

How it works

The converter takes in a JSON object response (see examples in the ./test folder.) Depending on which API you use, the converter will turn that into a shape that can be handled by the webvtt and srt scripts.

You provide the JSON object; then select the converter needed such as DeepgramConverter, WhisperTimestampedConverter, AssemblyAIConverter and so on. (If the API you want to use is not supported, please reach out to devrel@deepgram.com and we will do our best to add it.)

WebVTT from Deepgram Transcriptions

from deepgram_captions import DeepgramConverter, webvtt

transcription = DeepgramConverter(dg_response)
captions = webvtt(transcription)

SRT from Deepgram Transcriptions

from deepgram_captions import DeepgramConverter, srt

transcription = DeepgramConverter(dg_response)
captions = srt(transcription)

Line length

Add an optional integer parameter to set the line length of the caption.

line_length = 10

deepgram = DeepgramConverter(dg_speakers)
captions = webvtt(deepgram, line_length)

Other Converters

Whisper

Open AI's Whisper (through their API) does not provide timestamps, so a JSON response directly from OpenAI cannot be used with this package. However, there are a couple other options you can try:

Deepgram's Whisper Cloud

Use Deepgram's fully hosted Whisper Cloud, which gives you Whisper transcriptions along with the features that come with Deepgram's API such as timestamps. Use model=whisper when you make your request to Deepgram. Then use the DeepgramConverter to create the captions.

from deepgram_captions import DeepgramConverter, srt

transcription = DeepgramConverter(whisper_response)
captions = srt(transcription)

Whisper Timestamped

Whisper Timestamped adds word-level timestamps to OpenAI's Whisper speech-to-text transcriptions. Word-level timestamps are required for this package to create captions, which is why we have created the captions converter for Whisper Timestamped (and not OpenAI's Whisper).

from deepgram_captions import WhisperTimestampedConverter, webvtt

transcription = WhisperTimestampedConverter(whisper_response)
captions = webvtt(transcription)

Assembly AI

AssemblyAI is another popular speech-to-text API.

from deepgram_captions import AssemblyAIConverter, webvtt

transcription = AssemblyAIConverter(assembly_response)
captions = webvtt(transcription)

Output

Output WebVTT

When transcribing https://dpgr.am/spacewalk.wav, and running it through our library, this is the WebVTT output.

from deepgram_captions.converters import DeepgramConverter
from deepgram_captions.webvtt import webvtt

transcription = DeepgramConverter(dg_response)
captions = webvtt(transcription)
print(captions)

This is the result:

WEBVTT

NOTE
Transcription provided by Deepgram
Request Id: 686278aa-d315-4aeb-b2a9-713615544366
Created: 2023-10-27T15:35:56.637Z
Duration: 25.933313
Channels: 1

00:00:00.080 --> 00:00:03.220
Yeah. As as much as, it's worth celebrating,

00:00:04.400 --> 00:00:05.779
the first, spacewalk,

00:00:06.319 --> 00:00:07.859
with an all female team,

00:00:08.475 --> 00:00:10.715
I think many of us are looking forward

00:00:10.715 --> 00:00:13.215
to it just being normal and

00:00:13.835 --> 00:00:16.480
I think if it signifies anything, It is

00:00:16.779 --> 00:00:18.700
to honor the the women who came before

00:00:18.700 --> 00:00:21.680
us who, were skilled and qualified,

00:00:22.300 --> 00:00:24.779
and didn't get the same opportunities that we

00:00:24.779 --> 00:00:25.439
have today.

Output SRT

When transcribing https://dpgr.am/spacewalk.wav, and running it through our library, this is the SRT output.

from deepgram_captions import DeepgramConverter, srt

transcription = DeepgramConverter(dg_response)
captions = srt(transcription)
print(captions)

This is the result:

1
00:00:00,080 --> 00:00:03,220
Yeah. As as much as, it's worth celebrating,

2
00:00:04,400 --> 00:00:07,859
the first, spacewalk, with an all female team,

3
00:00:08,475 --> 00:00:10,715
I think many of us are looking forward

4
00:00:10,715 --> 00:00:14,235
to it just being normal and I think

5
00:00:14,235 --> 00:00:17,340
if it signifies anything, It is to honor

6
00:00:17,340 --> 00:00:19,820
the the women who came before us who,

7
00:00:20,140 --> 00:00:23,580
were skilled and qualified, and didn't get the

8
00:00:23,580 --> 00:00:25,439
same opportunities that we have today.

Documentation

You can learn more about the Deepgram API at developers.deepgram.com.

Development and Contributing

Interested in contributing? We ❤️ pull requests!

To make sure our community is safe for all, be sure to review and agree to our Code of Conduct. Then see the Contribution guidelines for more information.

Getting Help

We love to hear from you so if you have questions, comments or find a bug in the project, let us know! You can either:

deepgram / deepgram-python-captions

readme