Towards Orthographically-Informed Evaluation of Speech Recognition Systems for Indian Languages

• URL: http://arxiv.org/abs/2603.00941v1
• Date: Sun, 01 Mar 2026 06:20:35 GMT
• Title: Towards Orthographically-Informed Evaluation of Speech Recognition Systems for Indian Languages
• Abstract summary: evaluating ASR systems for Indian languages is challenging due to spelling variations.<n>Traditional Word Error Rate (WER) often presents a bleaker picture of system performance than what human users perceive.<n>We propose a framework for creating benchmarks that capture permissible variations.
• Score: 17.479176344686316
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Evaluating ASR systems for Indian languages is challenging due to spelling variations, suffix splitting flexibility, and non-standard spellings in code-mixed words. Traditional Word Error Rate (WER) often presents a bleaker picture of system performance than what human users perceive. Better aligning evaluation with real-world performance requires capturing permissible orthographic variations, which is extremely challenging for under-resourced Indian languages. Leveraging recent advances in LLMs, we propose a framework for creating benchmarks that capture permissible variations. Through extensive experiments, we demonstrate that OIWER, by accounting for orthographic variations, reduces pessimistic error rates (an average improvement of 6.3 points), narrows inflated model gaps (e.g., Gemini-Canary performance difference drops from 18.1 to 11.5 points), and aligns more closely with human perception than prior methods like WER-SN by 4.9 points.

Related papers

• Translation or Recitation? Calibrating Evaluation Scores for Machine Translation of Extremely Low-Resource Languages [39.985923582735936]
  We introduce the FRED Difficulty Metrics, which include the Fertility Ratio (F), Retrieval Proxy (R), Pre-training Exposure (E) and Corpus Diversity (D)<n>These metrics reveal that a significant portion of result variability is explained by train-test overlap and pre-training exposure rather than model capability.
  arXiv  Detail & Related papers  (2026-03-26T09:20:17Z)
• Benchmarking Bengali Dialectal Bias: A Multi-Stage Framework Integrating RAG-Based Translation and Human-Augmented RLAIF [0.3227658251731014]
  Large language models (LLMs) frequently exhibit performance biases against regional dialects of low-resource languages.<n>We propose a two-phase framework to evaluate dialectal bias in LLM question-answering across nine Bengali dialects.
  arXiv  Detail & Related papers  (2026-03-22T18:44:57Z)
• Crosslingual Optimized Metric for Translation Assessment of Indian Languages [3.3904531496305683]
  We create a large human evaluation ratings dataset for 13 Indian languages covering 21 translation directions.<n>We then train a neural translation evaluation metric named Cross-lingual Optimized Metric for Translation Assessment of Indian Languages (COMTAIL) on this dataset.<n>The best performing metric variants show significant performance gains over previous state-of-the-art when adjudging translation pairs with at least one Indian language.
  arXiv  Detail & Related papers  (2025-09-22T12:11:42Z)
• Out of Style: RAG's Fragility to Linguistic Variation [29.59506089890902]
  User queries exhibit greater linguistic variations and can trigger cascading errors across interdependent RAG components.<n>We analyze how varying four linguistic dimensions (formality, readability, politeness, and grammatical correctness) impact RAG performance.
  arXiv  Detail & Related papers  (2025-04-11T03:30:26Z)
• Advocating Character Error Rate for Multilingual ASR Evaluation [1.2597747768235845]
  We document the limitations of the word error rate (WER) as an evaluation metric and advocate for the character error rate (CER) as the primary metric. We show that CER avoids many of the challenges WER faces and exhibits greater consistency across writing systems. Our findings suggest that CER should be prioritized, or at least supplemented, in multilingual ASR evaluations to account for the varying linguistic characteristics of different languages.
  arXiv  Detail & Related papers  (2024-10-09T19:57:07Z)
• What is lost in Normalization? Exploring Pitfalls in Multilingual ASR Model Evaluations [0.0]
  We investigate the text normalization routine employed by leading ASR models, including OpenAI Whisper, Meta's MMS, Seamless, and Assembly AI's Conformer. Our research reveals that current text normalization practices, while aiming to standardize ASR outputs for fair comparison, are fundamentally flawed when applied to Indic scripts. We propose a shift towards developing text normalization routines that leverage native linguistic expertise.
  arXiv  Detail & Related papers  (2024-09-04T05:08:23Z)
• Machine Translation Meta Evaluation through Translation Accuracy Challenge Sets [92.38654521870444]
  We introduce ACES, a contrastive challenge set spanning 146 language pairs. This dataset aims to discover whether metrics can identify 68 translation accuracy errors. We conduct a large-scale study by benchmarking ACES on 50 metrics submitted to the WMT 2022 and 2023 metrics shared tasks.
  arXiv  Detail & Related papers  (2024-01-29T17:17:42Z)
• Unlikelihood Tuning on Negative Samples Amazingly Improves Zero-Shot Translation [79.96416609433724]
  Zero-shot translation (ZST) aims to translate between unseen language pairs in training data. The common practice to guide the zero-shot language mapping during inference is to deliberately insert the source and target language IDs. Recent studies have shown that language IDs sometimes fail to navigate the ZST task, making them suffer from the off-target problem.
  arXiv  Detail & Related papers  (2023-09-28T17:02:36Z)
• HyPoradise: An Open Baseline for Generative Speech Recognition with Large Language Models [81.56455625624041]
  We introduce the first open-source benchmark to utilize external large language models (LLMs) for ASR error correction. The proposed benchmark contains a novel dataset, HyPoradise (HP), encompassing more than 334,000 pairs of N-best hypotheses. LLMs with reasonable prompt and its generative capability can even correct those tokens that are missing in N-best list.
  arXiv  Detail & Related papers  (2023-09-27T14:44:10Z)
• Analyzing and Reducing the Performance Gap in Cross-Lingual Transfer with Fine-tuning Slow and Fast [50.19681990847589]
  Existing research has shown that a multilingual pre-trained language model fine-tuned with one (source) language also performs well on downstream tasks for non-source languages. This paper analyzes the fine-tuning process, discovers when the performance gap changes and identifies which network weights affect the overall performance most.
  arXiv  Detail & Related papers  (2023-05-19T06:04:21Z)
• Towards preserving word order importance through Forced Invalidation [80.33036864442182]
  We show that pre-trained language models are insensitive to word order. We propose Forced Invalidation to help preserve the importance of word order. Our experiments demonstrate that Forced Invalidation significantly improves the sensitivity of the models to word order.
  arXiv  Detail & Related papers  (2023-04-11T13:42:10Z)
• Retrieval-based Disentangled Representation Learning with Natural Language Supervision [61.75109410513864]
  We present Vocabulary Disentangled Retrieval (VDR), a retrieval-based framework that harnesses natural language as proxies of the underlying data variation to drive disentangled representation learning. Our approach employ a bi-encoder model to represent both data and natural language in a vocabulary space, enabling the model to distinguish intrinsic dimensions that capture characteristics within data through its natural language counterpart, thus disentanglement.
  arXiv  Detail & Related papers  (2022-12-15T10:20:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.