Papers › TASTE: Text-Aligned Speech Tokenization and Embedding for Spoken Language Modeling

TASTE: Text-Aligned Speech Tokenization and Embedding for Spoken Language Modeling

9 Apr 2025arXiv:2504.07053archive 2025-07-28

Liang-Hsuan Tseng, Yi-Chang Chen, Kuan-Yi Lee, Da-Shan Shiu, Hung-Yi Lee

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.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Syntology holds a reference implementation of TASTE, generated from the paper text (not the authors' code), at level 2 on Syntology's ladder: V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)” (quoted from Syntology's server card). No rung compares this code against the authors' implementation. For agents: get_reference_implementation(name='TASTE') (how to connect); the free check have('2504.07053') needs no token.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 20 code samples harvested from 2 repositories linked to this paper: 14 ran (of which 0 constructed an object rather than computing a result; 12 with no instrument failure: 0 honoured, 0 violated, 12 with no contract checked; 2 where Syntology's instrument failed) · 6 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 18 samples from 1 repository, 13 ran (11 with no instrument failure; 2 where Syntology's instrument failed); community (archive-listed): 2 samples from 1 repository, 1 ran (1 with no instrument failure; 0 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

mtkresearch/taste-spokenlm officialmentioned in papermentioned on GitHubpytorchSyntology: 18 samples, 13 ran (2 instrument-failed) · 5 unverified report
slp-rl/salmon mentioned on GitHubpytorchSyntology: 2 samples, 1 ran (0 instrument-failed) · 1 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

20 samples harvested: 14 ran (of which 0 constructed an object rather than computing a result; 12 with no instrument failure: 0 honoured, 0 violated, 12 with no contract checked; 2 where Syntology's instrument failed) · 6 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · our draft was wrong
12ran
6unverified

Licence: 20 of the 20 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: 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. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

mtkresearch/taste-spokenlm official repository 18 samples, 13 ran (2 instrument-failed) · 5 unverified
ContinueLatentLinearLastExtract taste_speech/modeling_taste.py ran no licence file found · pointer only · 75a8f67de68f1033 · report fixture: signature outputs:MockOutputs/0/?, vq_module:callable/0/?, audio_embeds:NoneType/0/? · returned tuple · level recorded 2026-09-24
ContinueLatentWeightedLayerExtract taste_speech/modeling_taste.py ran no licence file found · pointer only · ae3973909b9cc69d · report fixture: signature outputs:MockOutputs/0/?, vq_module:callable/0/?, audio_embeds:NoneType/0/? · returned tuple · level recorded 2026-09-24
ContinueWeightedLayerExtract taste_speech/modeling_taste.py ran no licence file found · pointer only · 93a5df0f3583bda3 · report fixture: signature outputs:MockOutputs/0/?, vq_module:callable/0/?, audio_embeds:NoneType/0/? · returned tuple · level recorded 2026-09-24
LinearLastExtract taste_speech/modeling_taste.py ran no licence file found · pointer only · 5fa309a42b342081 · report fixture: signature outputs:MockOutputs/0/?, vq_module:NoneType/0/?, audio_embeds:NoneType/0/? · returned tuple · level recorded 2026-09-24
MultiLinearLastExtract taste_speech/modeling_taste.py ran no licence file found · pointer only · cd4db8b78a432337 · report fixture: signature outputs:MockOutputs/0/?, vq_module:MockVQModule/0/?, audio_embeds:NoneType/0/? · returned tuple · level recorded 2026-09-24
ReferenceMixExtract taste_speech/modeling_taste.py ran no licence file found · pointer only · 279d6325801826ca · report fixture: signature outputs:obj/0/?, vq_module:NoneType/0/?, audio_embeds:Tensor/3/float · returned tuple · level recorded 2026-09-24
ReferenceMixFusion taste_speech/modeling_taste.py ran · metamorphic tier: invariant no licence file found · pointer only · a6b59d8ea035592f · report fixture: signature text_embeds:Tensor/2/float, audio_embeds:Tensor/2/float · returned Tensor · level recorded 2026-09-22
SimpleSumFusion taste_speech/modeling_taste.py ran · metamorphic tier: invariant no licence file found · pointer only · 8c00a209ba048c71 · report fixture: signature text_embeds:Tensor/2/float, audio_embeds:Tensor/2/float · returned Tensor · level recorded 2026-09-22
WeightedLayerExtract taste_speech/modeling_taste.py ran no licence file found · pointer only · b18446ab487a5f54 · report fixture: signature outputs:MockOutputs/0/?, vq_module:NoneType/0/?, audio_embeds:NoneType/0/? · returned tuple · level recorded 2026-09-24
WeightedSumFusion taste_speech/modeling_taste.py ran · metamorphic tier: invariant no licence file found · pointer only · 44ba0c56534073ad · report fixture: signature text_embeds:Tensor/2/float, audio_embeds:Tensor/2/float · returned Tensor · level recorded 2026-09-22
_AdaptedFFN taste_speech/modeling_taste.py ran · metamorphic tier: invariant no licence file found · pointer only · 4b94786aab62bee5 · report fixture: signature based_embeds:Tensor/2/float, ref_embeds:Tensor/2/float · returned Tensor · level recorded 2026-09-22
_find_all_linear_names taste_speech/modeling_taste.py ran · our draft was wrong no licence file found · pointer only · 5a098c87a1cf0f1d · report fixture: signature model:callable/0/? · returned list · level recorded 2026-09-22
generate_mask_from_length taste_speech/modeling_taste.py ran · our draft was wrong fingerprinted no licence file found · pointer only · d761c1c1939543f8 · report fixture: signature lengths:Tensor/1/int · returned Tensor · level recorded 2026-09-22
ExtractBase taste_speech/modeling_taste.py unverified no licence file found · pointer only · da4d4bf12c4d6aba · report
FusionBase taste_speech/modeling_taste.py unverified no licence file found · pointer only · 60bc29494edfd229 · report
LinearAllConcatExtract taste_speech/modeling_taste.py unverified no licence file found · pointer only · fb3b92de371e693a · report
TasteSampler taste_speech/modeling_taste.py unverified no licence file found · pointer only · e41a663f73152972 · report
TasteSpokenLM taste_speech/modeling_taste.py unverified no licence file found · pointer only · 7e00164c4e64dea2 · report
slp-rl/salmon community (archive-listed) 2 samples, 1 ran (0 instrument-failed) · 1 unverified
SpeechTokenizer baselines/slm_tokenizers.py ran no licence file found · pointer only · d9437193f548f512 · report fixture: signature x:list/1/Tensor, offset:int/0/int · returned list · level recorded 2026-09-24
get_gslm_speech_encoder baselines/slm_tokenizers.py unverified no licence file found · pointer only · 0566c277d0204a72 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. For a generated assembly, an independent peer is a second implementation generated separately from the same paper text; neither implementation read the authors' code. At level 2 the two agreed numerically on the fixture's inputs; level 3 is recorded when property tests derived from the paper also passed on both, and each level 3 card prints how many its report lists. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. A generated assembly's label was set when its code was generated for the method, and no model judged it; its card prints it as "by construction" or "not judged by a model". Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another. Where an assembly's phrase has no constructed figure, a brick that ran is not among the harvested samples this build reads, and whether a brick's run only constructed an object is read from that brick's harvested sample, so the figure was not counted; it does not say there were none.

3 assemblies on this paper: 2 harvested, 1 generated, listed in 2 groups (siblings lifted in one batch form one group).

generated Assembly for TASTE (graph method: TASTE), generated from the paper text, not the repository's code. 1 brick; individually on a fixture: 1 ran (1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. Ran end-to-end against an independent peer (level 2).

label matches by construction: the code was generated for this method (not judged by a model)

  1. align_word_level_embedding_with_llm entry · level 2 · ran · c07bfd297698765d3475635f42e241d2adbe110903a8f95351153b4982126f6f

2 sibling assemblies for TASTE (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for TASTE (graph method: TASTE). 18 bricks; individually on a fixture: 13 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 2 where Syntology's instrument failed) · 5 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. TasteSpokenLM entry · level 0 · unverified · 7e00164c4e64dea2
  2. TasteSampler helper · level 0 · unverified · e41a663f73152972
  3. _find_all_linear_names helper · level 1 · ran · our draft was wrong · 5a098c87a1cf0f1d
  4. generate_mask_from_length helper · level 1 · ran · our draft was wrong · d761c1c1939543f8
  5. ExtractBase helper · level 0 · unverified · da4d4bf12c4d6aba
  6. FusionBase helper · level 0 · unverified · 60bc29494edfd229
  7. LinearAllConcatExtract helper · level 0 · unverified · fb3b92de371e693a
  8. LinearLastExtract helper · level 1 · ran · 5fa309a42b342081
  9. MultiLinearLastExtract helper · level 1 · ran · cd4db8b78a432337
  10. ReferenceMixExtract helper · level 1 · ran · 279d6325801826ca
  11. ReferenceMixFusion helper · level 1 · ran · a6b59d8ea035592f
  12. SimpleSumFusion helper · level 1 · ran · 8c00a209ba048c71
  13. WeightedLayerExtract helper · level 1 · ran · b18446ab487a5f54
  14. WeightedSumFusion helper · level 1 · ran · 44ba0c56534073ad
  15. _AdaptedFFN helper · level 1 · ran · 4b94786aab62bee5
  16. ContinueLatentLinearLastExtract helper · level 1 · ran · 75a8f67de68f1033
  17. ContinueLatentWeightedLayerExtract model · level 1 · ran · ae3973909b9cc69d
  18. ContinueWeightedLayerExtract helper · level 1 · ran · 93a5df0f3583bda3

harvested Assembly for TASTE (graph method: TASTE). 2 bricks; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. SpeechTokenizer entry · level 1 · ran · d9437193f548f512
  2. get_gslm_speech_encoder helper · level 0 · unverified · 0566c277d0204a72

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2504.07053") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Language ModelingLanguage ModellingSpeech Tokenizationparameter-efficient fine-tuning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections