{"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/taste-text-aligned-speech-tokenization-and","title":"TASTE: Text-Aligned Speech Tokenization and Embedding for Spoken Language Modeling","arxiv_id":"2504.07053","date":"2025-04-09","proceeding":null,"authors":["Liang-Hsuan Tseng","Yi-Chang Chen","Kuan-Yi Lee","Da-Shan Shiu","Hung-Yi Lee"],"abstract":"Recent efforts target spoken language models (SLMs) that not only listen but also speak for more natural human-LLM interaction. Joint speech-text modeling is a promising direction to achieve this. However, the effectiveness of recent speech tokens for joint modeling remains underexplored. To address this, we introduce Text-Aligned Speech Tokenization and Embedding (TASTE), a method that directly addresses the modality gap by aligning speech token with the corresponding text transcription during the tokenization stage. We propose a method that can achieve this through a attention-based aggregation mechanism and with speech reconstruction as the training objective. We conduct extensive experiments and show that TASTE can preserve essential paralinguistic information while dramatically reducing the token sequence length. With TASTE, we perform straightforward joint spoken language modeling by using Low-Rank Adaptation on the pre-trained text LLM. Experimental results show that TASTE-based SLMs perform comparable to previous work on SALMON and StoryCloze; while significantly outperform other pre-trained SLMs on speech continuation across subjective and objective evaluations. To our knowledge, TASTE is the first end-to-end approach that utilizes a reconstruction objective to automatically learn a text-aligned speech tokenization and embedding suitable for spoken language modeling. Our demo, code, and model are available at https://mtkresearch.github.io/TASTE-SpokenLM.github.io.","url_abs":"https://arxiv.org/abs/2504.07053v2","url_pdf":"https://arxiv.org/pdf/2504.07053v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"taste-text-aligned-speech-tokenization-and","repo_url":"https://github.com/mtkresearch/taste-spokenlm","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"taste-text-aligned-speech-tokenization-and","repo_url":"https://github.com/slp-rl/salmon","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"speech-tokenization","task_name":"Speech Tokenization"},{"task_slug":"parameter-efficient-fine-tuning","task_name":"parameter-efficient fine-tuning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2504.07053","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.07053"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+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":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/slp-rl/salmon","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mtkresearch/taste-spokenlm","reach":null}],"summary":{"ran":12,"ran_draft_wrong":2,"unverified":6},"by_repo_kind":{"official":{"samples":18,"ran":13,"repositories":1},"listed":{"samples":2,"ran":1,"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":20,"samples":[{"code_sha256_prefix":"75a8f67de68f1033","entry":"ContinueLatentLinearLastExtract","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"75a8f67de68f1033"}},{"code_sha256_prefix":"ae3973909b9cc69d","entry":"ContinueLatentWeightedLayerExtract","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ae3973909b9cc69d"}},{"code_sha256_prefix":"93a5df0f3583bda3","entry":"ContinueWeightedLayerExtract","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"93a5df0f3583bda3"}},{"code_sha256_prefix":"5fa309a42b342081","entry":"LinearLastExtract","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5fa309a42b342081"}},{"code_sha256_prefix":"cd4db8b78a432337","entry":"MultiLinearLastExtract","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"cd4db8b78a432337"}},{"code_sha256_prefix":"279d6325801826ca","entry":"ReferenceMixExtract","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"279d6325801826ca"}},{"code_sha256_prefix":"a6b59d8ea035592f","entry":"ReferenceMixFusion","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a6b59d8ea035592f"}},{"code_sha256_prefix":"8c00a209ba048c71","entry":"SimpleSumFusion","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8c00a209ba048c71"}},{"code_sha256_prefix":"d9437193f548f512","entry":"SpeechTokenizer","repo":"slp-rl/salmon","repo_kind":"listed","path":"baselines/slm_tokenizers.py","file_url":"https://github.com/slp-rl/salmon/blob/HEAD/baselines/slm_tokenizers.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d9437193f548f512"}},{"code_sha256_prefix":"b18446ab487a5f54","entry":"WeightedLayerExtract","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"b18446ab487a5f54"}},{"code_sha256_prefix":"44ba0c56534073ad","entry":"WeightedSumFusion","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"44ba0c56534073ad"}},{"code_sha256_prefix":"4b94786aab62bee5","entry":"_AdaptedFFN","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4b94786aab62bee5"}},{"code_sha256_prefix":"5a098c87a1cf0f1d","entry":"_find_all_linear_names","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5a098c87a1cf0f1d"}},{"code_sha256_prefix":"d761c1c1939543f8","entry":"generate_mask_from_length","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d761c1c1939543f8"}},{"code_sha256_prefix":"da4d4bf12c4d6aba","entry":"ExtractBase","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"da4d4bf12c4d6aba"}},{"code_sha256_prefix":"60bc29494edfd229","entry":"FusionBase","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"60bc29494edfd229"}},{"code_sha256_prefix":"fb3b92de371e693a","entry":"LinearAllConcatExtract","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"fb3b92de371e693a"}},{"code_sha256_prefix":"e41a663f73152972","entry":"TasteSampler","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"e41a663f73152972"}},{"code_sha256_prefix":"7e00164c4e64dea2","entry":"TasteSpokenLM","repo":"mtkresearch/taste-spokenlm","repo_kind":"official","path":"taste_speech/modeling_taste.py","file_url":"https://github.com/mtkresearch/taste-spokenlm/blob/HEAD/taste_speech/modeling_taste.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"7e00164c4e64dea2"}},{"code_sha256_prefix":"0566c277d0204a72","entry":"get_gslm_speech_encoder","repo":"slp-rl/salmon","repo_kind":"listed","path":"baselines/slm_tokenizers.py","file_url":"https://github.com/slp-rl/salmon/blob/HEAD/baselines/slm_tokenizers.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0566c277d0204a72"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}