sabuhigr / sabuhi-model

Whisper AI with channel separation and speaker diarization

  • Public
  • 25.5K runs
  • L40S
Run with an API
Playground API Examples README Versions

Input

Current Time 00:00:000
Duration 00:00:000
Loaded: 0%
Stream Type LIVE
Remaining Time 00:00:000
 
1x
*file

Audio file

*string
Shift + Return to add a new line

Your Hugging Face token for speaker diarization

string

Choose a Whisper model.

Default: "large-v2"

string

Choose the format for the transcription

Default: "plain text"

boolean

Translate the text to English when set to True

Default: false

*string

language spoken in the audio, specify None to perform language detection

number

temperature to use for sampling

Default: 0

number

optional patience value to use in beam decoding, as in https://arxiv.org/abs/2204.05424, the default (1.0) is equivalent to conventional beam search

string
Shift + Return to add a new line

comma-separated list of token ids to suppress during sampling; '-1' will suppress most special characters except common punctuations

Default: "-1"

string
Shift + Return to add a new line

optional text to provide as a prompt for the first window.

boolean

if True, provide the previous output of the model as a prompt for the next window; disabling may make the text inconsistent across windows, but the model becomes less prone to getting stuck in a failure loop

Default: true

number

temperature to increase when falling back when the decoding fails to meet either of the thresholds below

Default: 0.2

number

if the gzip compression ratio is higher than this value, treat the decoding as failed

Default: 2.4

number

if the average log probability is lower than this value, treat the decoding as failed

Default: -1

number

if the probability of the <|nospeech|> token is higher than this value AND the decoding has failed due to `logprob_threshold`, consider the segment as silence

Default: 0.6

integer

Select 2 if record is stereo, 1 if is mono.Default is 1 for mono records

Default: 1

integer

Select 2 if record is stereo, 1 if is mono.Default is 1 for mono records

Default: 1

Add a payment method to run this model.

Sign in with GitHub

By signing in, you agree to our
terms of service and privacy policy

Output

segments

[ { "end": 3.98, "text": " تقع القاهرة على جوانب جزر نهر النيل في شمال مصر", "start": 0.06, "words": [ { "end": 0.422, "word": "تقع", "score": 0.884, "start": 0.06 }, { "end": 1.025, "word": "القاهرة", "score": 0.899, "start": 0.442, "speaker": "SPEAKER_00" }, { "end": 1.306, "word": "على", "score": 0.874, "start": 1.085, "speaker": "SPEAKER_00" }, { "end": 1.95, "word": "جوانب", "score": 0.963, "start": 1.367, "speaker": "SPEAKER_00" }, { "end": 2.372, "word": "جزر", "score": 0.928, "start": 2.01, "speaker": "SPEAKER_00" }, { "end": 2.713, "word": "نهر", "score": 0.991, "start": 2.412, "speaker": "SPEAKER_00" }, { "end": 3.015, "word": "النيل", "score": 0.527, "start": 2.734, "speaker": "SPEAKER_00" }, { "end": 3.196, "word": "في", "score": 0.999, "start": 3.075, "speaker": "SPEAKER_00" }, { "end": 3.678, "word": "شمال", "score": 0.986, "start": 3.276, "speaker": "SPEAKER_00" }, { "end": 3.98, "word": "مصر", "score": 0.534, "start": 3.759, "speaker": "SPEAKER_00" } ], "speaker": "SPEAKER_00" } ]

transcription

[{'start': 0.06, 'end': 3.98, 'text': ' تقع القاهرة على جوانب جزر نهر النيل في شمال مصر', 'words': [{'word': 'تقع', 'start': 0.06, 'end': 0.422, 'score': 0.884}, {'word': 'القاهرة', 'start': 0.442, 'end': 1.025, 'score': 0.899, 'speaker': 'SPEAKER_00'}, {'word': 'على', 'start': 1.085, 'end': 1.306, 'score': 0.874, 'speaker': 'SPEAKER_00'}, {'word': 'جوانب', 'start': 1.367, 'end': 1.95, 'score': 0.963, 'speaker': 'SPEAKER_00'}, {'word': 'جزر', 'start': 2.01, 'end': 2.372, 'score': 0.928, 'speaker': 'SPEAKER_00'}, {'word': 'نهر', 'start': 2.412, 'end': 2.713, 'score': 0.991, 'speaker': 'SPEAKER_00'}, {'word': 'النيل', 'start': 2.734, 'end': 3.015, 'score': 0.527, 'speaker': 'SPEAKER_00'}, {'word': 'في', 'start': 3.075, 'end': 3.196, 'score': 0.999, 'speaker': 'SPEAKER_00'}, {'word': 'شمال', 'start': 3.276, 'end': 3.678, 'score': 0.986, 'speaker': 'SPEAKER_00'}, {'word': 'مصر', 'start': 3.759, 'end': 3.98, 'score': 0.534, 'speaker': 'SPEAKER_00'}], 'speaker': 'SPEAKER_00'}]

detected_language

arabic
Generated in
Tweak itReportView full prediction

This output was created using a different version of the model, sabuhigr/sabuhi-model:48c9b65d.

Run time and cost

This model costs approximately $0.055 to run on Replicate, or 18 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia L40S GPU hardware. Predictions typically complete within 57 seconds. The predict time for this model varies significantly based on the inputs.

Readme

This model doesn't have a readme.