OpenTrain AI
No verified implementation yetPretrained Models Available

Robust Long-Form Bangla Speech Processing: Automatic Speech Recognition and Speaker Diarization

MD. Sagor Chowdhury, Adiba Fairooz Chowdhury

February 25, 2026arXiv: 2602.21741
0 repos~a few days to reproduce
arXiv PDF

Abstract

We describe our end-to-end system for Bengali long-form speech recognition (ASR) and speaker diarization submitted to the DL Sprint 4.0 competition on Kaggle. Bengali presents substantial challenges for both tasks: a large phoneme inventory, significant dialectal variation, frequent code-mixing with English, and a relative scarcity of large-scale labelled corpora. For ASR we achieve a best private Word Error Rate (WE...

Results & Benchmarks

TaskDatasetMetricValue
DetectionSeamlessM4T (fixed)Private WER.1.1502
DetectionSeamlessM4T (VAD)Public WER0.7435
Computer visionTugstugiWER34.07
Computer visionOurs (ASR)WER37.74

Hardware Requirements

  • Expect multi-day setup/compute for meaningful reproduction based on current guidance.

Best Implementation

Maintained implementation evidence is not confirmed for this paper yet.

Use the Implementation Status and Reproduction Path sections below for the current action plan.

Reproduction Path

Follow this baseline workflow to decide if this paper is worth immediate prototyping.

  1. 1

    Use the paper and benchmark evidence to scope a baseline reproduction plan.

  2. 2

    Track assumptions and missing details in an experiment log before coding.

Time to first repro: a few daysEstimate is based on paper-only reproduction flow

Additional Implementations

No additional verified repositories beyond the primary recommendation.

Hugging Face Artifacts

No direct paper-linked artifacts were found. Showing strongest curated related artifacts.

Curated Related