← Back to all news
arXiv Computation and Language By Hiwa Asadpour

Unadapted Multilingual ASR on a Garrusi Kurdish Evaluation Set: A Common-Reference Staged Normalization Analysis

Read the original on arXiv Computation and Language →

The paper evaluates a multilingual ASR model (MMS‑1B‑all) on a Garrusi Kurdish dataset using a common‑reference staged normalization approach. By normalizing both reference and hypothesis, the authors show that raw Arabic‑script hypotheses yield a 111.70 % WER, which drops to 97.85 % after folding into a reduced orthography, highlighting the impact of orthographic differences on error measurement. A Southern Kurdish fine‑tuned system performs worse, and residual errors are partly due to scoring‑pipeline limitations rather than recognition failures.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

Related stories

arXiv Computation and Language
2d ago

ufakzeka-kulak: Building and Evaluating a 31.6M-Parameter Turkish Speech Recognizer from Scratch

ufakzeka-kulak is an open Turkish speech‑to‑text model with 31.6 M parameters, built using a FastConformer encoder, a token‑and‑duration transducer (TDT), and a CTC head. It was trained from scratch on 4,977 hours of data—281 hours of human transcription and 4,696 hours of machine‑generated subtitles that were verified by a second reading—under licenses permitting commercial use. The model outputs cased, punctuated Turkish text with numbers written as words, achieving a normalised WER of 0.0921 on FLEURS Turkish and 0.0916 on Common Voice 27.0 Turkish, outperforming Whisper base and small models, and it is released under Apache‑2.0.

By Sait Furkan Teke (ufak AI)
multimodal
More like this →
arXiv Computation and Language
Oct 22

How Far Can the Gemma 4 Speech Encoder Go Without Changing Its Weights? Saudi and Gulf Arabic Speech Recognition with a Frozen and an Adapter-Augmented Encoder

The paper evaluates a 305 M‑parameter audio encoder shared by Gemma 4 and EmbeddingGemma 2 for Saudi and Gulf Arabic speech recognition without altering its weights. Two approaches are compared: a frozen system that trains only a 22.6 M‑parameter CTC reader, and an adapted system that adds 3.1 M low‑rank adapters, a 41.5 M‑parameter hybrid CTC/attention reader, and joint decoding with a sub‑word language model. The adapted system matches or surpasses a 0.6 B‑parameter fine‑tuned recogniser on several datasets, while the frozen system remains competitive in character error rate and runs three‑quarters faster on CPU with greedy decoding.

By Amr Abdelkhalik
llmsfine-tuningmultimodal
More like this →
arXiv Computation and Language
Sep 11

Assessing the Reusability of Public Speech Resources for Low-Resource Languages: A Central Kurdish Case Study

The article reviews a recent Kurdish speech resource release that includes three voices, 35 hours of recorded speech, and a supporting paper. It finds several issues with the public files—mislisted equipment, unlabeled test data, and a coding error—while noting that the download page overstates the results and recommends a single voice for general use. The authors argue that these problems can largely be corrected with existing information, improving the resource’s usability for future work.

By Hiwa Asadpour
More like this →
arXiv Machine Learning
Oct 7

Making COMET Comparable Across Scripts: Diagnosis and Correction of Tokeniser-Induced Script Bias in Indic MT Evaluation

arXiv:2610.08159v1 Announce Type: cross Abstract: COMET reports translation quality as a single number, and that number is routinely compared across target languages written in different scripts. Suc...

By G. L. John Salvin (Indian Institute of Technology Palakkad), Swapnil Hingmire (Indian Institute of Technology Palakkad)
safety
More like this →