{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/arxiv-2601-01461","title":"Bridging the gap: A comparative exploration of Speech-LLM and end-to-end architecture for multilingual conversational ASR","arxiv_id":"2601.01461","date":"2026-01-04","proceeding":null,"authors":["Yuxiang Mei","Dongxing Xu","Jiaen Liang","Yanhua Long"],"abstract":"The INTERSPEECH 2025 Challenge on Multilingual Conversational Speech Language Models (MLC-SLM) promotes multilingual conversational ASR with large language models (LLMs). Our previous SHNU-mASR system adopted a competitive parallel-speech-encoder architecture that integrated Whisper and mHuBERT with an LLM. However, it faced two challenges: simple feature concatenation may not fully exploit complementary information, and the performance gap between LLM-based ASR and end-to-end(E2E) encoder-decoder ASR remained unexplored. In this work, we present an enhanced LLM-based ASR framework that combines fine-tuned Whisper and mHuBERT encoders with an LLM to enrich speech representations. We first evaluate E2E Whisper models with LoRA and full fine-tuning on the MLC-SLM ASR task, and then propose cross-attention-based fusion mechanisms for the parallel-speech-encoder. On the official evaluation set of the MLC-SLM Challenge, our system achieves a CER/WER of 10.69%, ranking on par with the top-ranked Track 1 systems, even though it uses only 1,500 hours of baseline training data compared with their large-scale training sets. Nonetheless, we find that our final LLM-based ASR still does not match the performance of a fine-tuned E2E Whisper model, providing valuable empirical guidance for future Speech-LLM design. Our code is publicly available at https://github.com/1535176727/MLC-SLM.","url_abs":"https://arxiv.org/abs/2601.01461","url_pdf":"https://arxiv.org/pdf/2601.01461","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2601.01461","atlas_url":null,"mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2601.01461"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/1535176727/MLC-SLM","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":5},"by_repo_kind":{"found_in_text":{"samples":5,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"838e2dc1ef2dd212","entry":"add_space_between_chars","repo":"1535176727/MLC-SLM","repo_kind":"found_in_text","path":"local/add_space_between_chars_asr.py","file_url":"https://github.com/1535176727/MLC-SLM/blob/HEAD/local/add_space_between_chars_asr.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"838e2dc1ef2dd212"}},{"code_sha256_prefix":"223e0d390d4487f8","entry":"normalize_text_lines","repo":"1535176727/MLC-SLM","repo_kind":"found_in_text","path":"local/Whisper_norm.py","file_url":"https://github.com/1535176727/MLC-SLM/blob/HEAD/local/Whisper_norm.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"223e0d390d4487f8"}},{"code_sha256_prefix":"8fef80a5b9af66eb","entry":"process_txt_files","repo":"1535176727/MLC-SLM","repo_kind":"found_in_text","path":"local/prepare_segments.py","file_url":"https://github.com/1535176727/MLC-SLM/blob/HEAD/local/prepare_segments.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8fef80a5b9af66eb"}},{"code_sha256_prefix":"b9bae46ad6f10bcb","entry":"process_txt_files","repo":"1535176727/MLC-SLM","repo_kind":"found_in_text","path":"local/prepare_test_segments.py","file_url":"https://github.com/1535176727/MLC-SLM/blob/HEAD/local/prepare_test_segments.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b9bae46ad6f10bcb"}},{"code_sha256_prefix":"87cc4b2ad4829a32","entry":"trim_repeated_suffix","repo":"1535176727/MLC-SLM","repo_kind":"found_in_text","path":"local/remove_repetition.py","file_url":"https://github.com/1535176727/MLC-SLM/blob/HEAD/local/remove_repetition.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"87cc4b2ad4829a32"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.CL","source":"arxiv_2026.jsonl"},"syntology_extracted_results":null}