BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pyconde-pydata-2026//speaker//8RV9AV
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-pyconde-pydata-2026-LYCBNT@pretalx.com
DTSTART;TZID=CET:20260414T151000
DTEND;TZID=CET:20260414T155500
DESCRIPTION:Building machine learning models for audio deepfake detection s
 eems straightforward until datasets span multiple languages\, such as Hind
 i\, Korean\, Mandarin\, and German. In practice\, multilingual Automatic S
 peech Recognition (ASR) systems often fail in production because language-
 specific acoustic variations and assumptions about the processing pipeline
  break down at scale.\n\nThis talk examines the engineering challenges of 
 building a multilingual deepfake detection system using a Python-centric p
 ipeline. It covers practical issues encountered during large-scale audio p
 reprocessing\, including memory-efficient data loading\, resumable feature
 -extraction workflows\, and validation strategies designed to prevent cros
 s-lingual leakage. The session also shares lessons from deploying a multil
 ingual ASR-based system\, with a focus on pipeline structure\, evaluation 
 correctness\, and operational robustness in real-world settings.
DTSTAMP:20260412T141741Z
LOCATION:Palladium [2nd Floor]
SUMMARY:What Breaks When Automatic Speech Recognition Systems Go Multilingu
 al - Rashmi Nagpal
URL:https://pretalx.com/pyconde-pydata-2026/talk/LYCBNT/
END:VEVENT
END:VCALENDAR