Reinforcement Learning for Data-Efficient Code-Switched ASR

• URL: http://arxiv.org/abs/2607.02757v1
• Date: Thu, 02 Jul 2026 20:44:53 GMT
• Title: Reinforcement Learning for Data-Efficient Code-Switched ASR
• Abstract summary: We propose a practical reinforcement learning with verifiable rewards recipe for data-efficient adaptation of audio-language models to code-switched ASR.<n>We show that RLVR with 10% of the data matches LoRA supervised fine-tuning trained on the full dataset, with the largest gains on typologically distant pairs.
• Score: 0.8594140167290097
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Audio-language models can be prompted for code-switched speech, but their decoding is not optimized for code-switching and often fails at language boundaries. We propose a practical reinforcement learning with verifiable rewards recipe for data-efficient adaptation of audio-language models to code-switched ASR using group relative policy optimization, combining an error rate reward with a script fidelity reward that penalizes wrong writing systems and a two-pass draft-and-refinement procedure. Using Qwen2-Audio as a reproducible testbed across 10 language pairs, training on only TTS code-switched speech, we show that RLVR with 10% of the data matches LoRA supervised fine-tuning trained on the full dataset, with the largest gains on typologically distant pairs. The error rate reward eliminates translation errors while the script fidelity reward separately reduces script contamination without degradation. These gains transfer zero-shot to a human-recorded code-switching corpus.

Related papers

• Light-weight Pronunciation Assessment via Discrete Speech Token Surprisal [10.584132982823883]
  Training automated pronunciation assessment often relies on labeled learner errors or non-native corpora that are costly to collect.<n>We propose a lightweight framework trained only on native speech resources, operating unsupervised or lightly calibrated with a small set of scored utterances.
  arXiv  Detail & Related papers  (2026-06-18T08:04:16Z)
• Hierarchical Policy Optimization for Simultaneous Translation of Unbounded Speech [50.45710815530982]
  Simultaneous speech translation (SST) generates translations while receiving partial speech input.<n>Recent advances show that large language models (LLMs) can substantially improve SST quality, but at the cost of high computational overhead.<n>We propose a Hierarchical Policy Optimization (HPO) approach that trains post-train models on imperfect SFT data.<n> Experiments on English to Chinese/German/Japanese demonstrate improvements of over +7 COMET score and +1.25 MetricX score at a latency of 1.5 seconds.
  arXiv  Detail & Related papers  (2026-04-22T19:43:51Z)
• Improving Code-Switching Speech Recognition with TTS Data Augmentation [58.34842693152991]
  This paper explores multilingual text-to-speech (TTS) models as an effective data augmentation technique to address this shortage.<n>We fine-tune the multilingual CosyVoice2 TTS model on the SEAME dataset to generate synthetic conversational Chinese-English code-switching speech.
  arXiv  Detail & Related papers  (2026-01-02T10:11:51Z)
• ATRI: Mitigating Multilingual Audio Text Retrieval Inconsistencies by Reducing Data Distribution Errors [42.469131776062724]
  Existing Multilingual audio-text retrieval schemes suffer from inconsistencies for instance similarity matching across languages.<n>We propose a consistent ML-ATR scheme using 1-to-k contrastive learning and audio-English co-anchor contrastive learning.<n>Our scheme achieves state-of-the-art performance on recall and consistency metrics for eight mainstream languages, including English.
  arXiv  Detail & Related papers  (2025-02-20T15:06:15Z)
• Multilingual self-supervised speech representations improve the speech recognition of low-resource African languages with codeswitching [65.74653592668743]
  Finetuning self-supervised multilingual representations reduces absolute word error rates by up to 20%. In circumstances with limited training data finetuning self-supervised representations is a better performing and viable solution.
  arXiv  Detail & Related papers  (2023-11-25T17:05:21Z)
• Speech collage: code-switched audio generation by collaging monolingual corpora [50.356820349870986]
  Speech Collage is a method that synthesizes CS data from monolingual corpora by splicing audio segments. We investigate the impact of generated data on speech recognition in two scenarios.
  arXiv  Detail & Related papers  (2023-09-27T14:17:53Z)
• Optimizing Bilingual Neural Transducer with Synthetic Code-switching Text Generation [10.650573361117669]
  Semi-supervised training and synthetic code-switched data can improve the bilingual ASR system on code-switching speech. Our final system achieves 25% mixed error rate (MER) on the ASCEND English/Mandarin code-switching test set.
  arXiv  Detail & Related papers  (2022-10-21T19:42:41Z)
• Pre-Training Transformer Decoder for End-to-End ASR Model with Unpaired Speech Data [145.95460945321253]
  We introduce two pre-training tasks for the encoder-decoder network using acoustic units, i.e., pseudo codes. The proposed Speech2C can relatively reduce the word error rate (WER) by 19.2% over the method without decoder pre-training.
  arXiv  Detail & Related papers  (2022-03-31T15:33:56Z)
• Lexically Aware Semi-Supervised Learning for OCR Post-Correction [90.54336622024299]
  Much of the existing linguistic data in many languages of the world is locked away in non-digitized books and documents. Previous work has demonstrated the utility of neural post-correction methods on recognition of less-well-resourced languages. We present a semi-supervised learning method that makes it possible to utilize raw images to improve performance.
  arXiv  Detail & Related papers  (2021-11-04T04:39:02Z)

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.