rafaelgalle/whisper-diarization-advanced

Official View on Replicate →

Ultra-fast, customizable speech-to-text and speaker diarization for noisy, multi-speaker audio. Includes advanced noise reduction, stereo channel support, and flexible audio preprocessing—ideal for call centers, meetings, and podcasts.

Capabilities

No capability data available

Cost

Community model (estimated from hardware time)

Input Parameters

Name	Type	Description	Default	Constraints
`file_path`	string (uri)	Audio file	`—`	—
`file_string`	string	Base64 audio	`—`	—
`file_url`	string	Direct URL	`—`	—
`highpass_freq`	integer	—	`45`	—
`language`	string	Language code like 'en', 'pt'	`—`	—
`lowpass_freq`	integer	—	`8000`	—
`num_speakers`	integer	Leave empty to autodetect	`—`	min: 1, max: 50
`preprocess`	integer	0=None, 1=Sanitize, 2=+Filter, 3=+ReduceNoise, 4=+Normalize	`0`	min: 0, max: 4
`prompt`	string	Names/acronyms, separated by punctuation	`—`	—
`prop_decrease`	number	—	`0.3`	—
`stationary`	boolean	—	`true`	—
`target_dBFS`	number	—	`-18`	—
`translate`	boolean	Translate to English	`false`	—

file_path string

Audio file

file_string string

Base64 audio

file_url string

Direct URL

highpass_freq integer

Default: 45

language string

Language code like 'en', 'pt'

lowpass_freq integer

Default: 8000

num_speakers integer

Leave empty to autodetect

min: 1, max: 50

preprocess integer

0=None, 1=Sanitize, 2=+Filter, 3=+ReduceNoise, 4=+Normalize

Default: 0 min: 0, max: 4

prompt string

Names/acronyms, separated by punctuation

prop_decrease number

Default: 0.3

stationary boolean

Default: true

target_dBFS number

Default: -18

translate boolean

Translate to English

Default: false

Version: 56dcb55b658e Updated: 2/26/2026 387.3K runs