Papers › Lyra: An Efficient and Speech-Centric Framework for Omni-Cognition

Lyra: An Efficient and Speech-Centric Framework for Omni-Cognition

12 Dec 2024arXiv:2412.09501archive 2025-07-28

Zhisheng Zhong, Chengyao Wang, Yuqi Liu, Senqiao Yang, Longxiang Tang, Yuechen Zhang, Jingyao Li, Tianyuan Qu, Yanwei Li, Yukang Chen, Shaozuo Yu, Sitong Wu, Eric Lo, Shu Liu, Jiaya Jia

As Multi-modal Large Language Models (MLLMs) evolve, expanding beyond single-domain capabilities is essential to meet the demands for more versatile and efficient AI. However, previous omni-models have insufficiently explored speech, neglecting its integration with multi-modality. We introduce Lyra, an efficient MLLM that enhances multimodal abilities, including advanced long-speech comprehension, sound understanding, cross-modality efficiency, and seamless speech interaction. To achieve efficiency and speech-centric capabilities, Lyra employs three strategies: (1) leveraging existing open-source large models and a proposed multi-modality LoRA to reduce training costs and data requirements; (2) using a latent multi-modality regularizer and extractor to strengthen the relationship between speech and other modalities, thereby enhancing model performance; and (3) constructing a high-quality, extensive dataset that includes 1.5M multi-modal (language, vision, audio) data samples and 12K long speech samples, enabling Lyra to handle complex long speech inputs and achieve more robust omni-cognition. Compared to other omni-methods, Lyra achieves state-of-the-art performance on various vision-language, vision-speech, and speech-language benchmarks, while also using fewer computational resources and less training data.

PaperPDFCodeCode Syntology ran

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.

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

Code

Syntology 19 code samples harvested from 1 repository linked to this paper: 14 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 1 violated, 7 with no contract checked; 6 where Syntology's instrument failed) · 5 unverified. Where Syntology's instrument failed: 5 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 19 samples from 1 repository, 14 ran (8 with no instrument failure; 6 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 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 2 of the 14 samples that ran have no recorded date.

dvlab-research/Lyra officialmentioned on GitHubpytorchApache-2.0Syntology: 19 samples, 14 ran (6 instrument-failed) · 5 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

19 samples harvested: 14 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 1 violated, 7 with no contract checked; 6 where Syntology's instrument failed) · 5 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · violated contract
5ran · our draft was wrong
1ran; the contract check could not run
7ran
5unverified

Licence: 4 of the 19 samples are pointer only, meaning Syntology does not serve that copy's text. For 4 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from dvlab-research/Lyra. “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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

dvlab-research/Lyra official repository 19 samples, 14 ran (6 instrument-failed) · 5 unverified
compute_softdtw in 9 papers lyra/model/soft_dtw_cuda.py ran Apache-2.0 (permissive) · 34f95bf62dff3ebd · report fixture: signature D:ndarray/3/float, gamma:float/0/float, bandwidth:int/0/int · returned ndarray · level recorded 2026-09-25
compute_softdtw_backward in 9 papers lyra/model/soft_dtw_cuda.py ran Apache-2.0 (permissive) · ec64f40bfc12544a · report fixture: signature D_:ndarray/3/float, R:ndarray/3/float, gamma:float/0/float, bandwidth:int/0/int · returned ndarray · level recorded 2026-09-25
extract_patches in 34 papers lyra/mm_utils.py ran; the contract check could not run (RAISES) · metamorphic tier: well formed Apache-2.0 (permissive) here; the stored copy is pointer only · b0a851cae92754ff · report fixture: signature image:Image/0/?, patch_size:int/0/int, overlap_ratio:float/0/float · returned list · level recorded 2026-09-22
forward_layer lyra/model/qwen2vl_top_attn.py ran Apache-2.0 (permissive) · aebe8894bde0f55e · report fixture: signature self:callable/0/?, hidden_states:Tensor/3/float, attention_mask:NoneType/0/?, position_ids:Tensor/2/int, past_key_value:NoneType/0/?, output_attentions:bool/0/bool, use_cache:bool/0/bool, cache_position:Tensor/1/int, position_embeddings:tuple/1/Tensor, context_labels:Tensor/2/bool, text_labels:Tensor/2/bool · returned tuple · level recorded 2026-09-25
forward_model lyra/model/qwen2vl_top_attn.py ran Apache-2.0 (permissive) · 74e2126c3223cdd8 · report fixture: signature self:callable/0/?, input_ids:Tensor/2/int, attention_mask:Tensor/2/float, position_ids:Tensor/2/int, past_key_values:PastKeyValues/0/?, inputs_embeds:NoneType/0/?, text_labels:Tensor/2/bool, context_labels:Tensor/2/bool, labels:Tensor/2/bool, use_cache:bool/0/bool, output_attentions:bool/0/bool, output_hidden_states:bool/0/bool, return_dict:bool/0/bool, cache_position:Tensor/1/int · returned CompressModelOutputWithPast · level recorded 2026-09-25
get_chunk in 197 papers lyra/eval/model_lyra_image_speech.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 42a46570620cd9fa · report fixture: signature lst:list/1/int, n:int/0/int, k:int/0/int · returned list · level recorded, date not recorded
img_to_base64 lyra/conversation.py ran · our draft was wrong Apache-2.0 (permissive) · f06df1307d1b0d63 · report fixture: signature img_file_path:str/0/str · returned str · level recorded 2026-09-25
initialize_input_ids_for_generation lyra/model/builder.py ran Apache-2.0 (permissive) · 91cfc0ec83571808 · report fixture: signature self:callable/0/?, inputs:Tensor/2/int, bos_token_id:Tensor/0/int, model_kwargs:dict/0/? · returned Tensor · level recorded 2026-09-25
mov_to_base64 lyra/conversation.py ran · our draft was wrong Apache-2.0 (permissive) · 34782af36cf765e7 · report fixture: signature mov_file_path:str/0/str · returned str · level recorded 2026-09-25
save_result in 10 papers lyra/eval/model_lyra_text_speech.py ran Apache-2.0 (permissive) · 7719de171ef38b85 · report fixture: signature result:list/1/dict, result_file:str/0/str, remove_duplicate:str/0/str · returned str · level recorded 2026-09-25
split_list in 223 papers lyra/eval/model_lyra_image_speech.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 076c252c52cbb161 · report fixture: signature lst:list/1/int, n:int/0/int · returned list · level recorded, date not recorded
timed_run lyra/model/soft_dtw_cuda.py ran Apache-2.0 (permissive) · 89a318a54beb022e · report fixture: signature a:Tensor/3/float, b:Tensor/3/float, sdtw:callable/0/? · returned tuple · level recorded 2026-09-25
violates_moderation in 59 papers lyra/utils.py ran · violated contract Apache-2.0 (permissive) here; the stored copy is pointer only · f9939a84b9a65279 · report fixture: signature text:str/0/str · returned bool · level recorded 2026-09-26
wav_to_base64 lyra/conversation.py ran · our draft was wrong Apache-2.0 (permissive) · dc99ac2d1a2f8f6b · report fixture: signature wav_file_path:str/0/str · returned str · level recorded 2026-09-25
auto_pad_images in 20 papers lyra/mm_utils.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · 3c8823948724e269 · report
build_unit_tokenizer lyra/serve/model_worker.py unverified Apache-2.0 (permissive) · 23e91c5c56d7ff80 · report
forward_attn lyra/model/qwen2vl_top_attn.py unverified Apache-2.0 (permissive) · 68283f6ff784cc32 · report
pretty_print_semaphore in 55 papers lyra/utils.py unverified Apache-2.0 (permissive) · 37899f22fb191b37 · report
resize_and_center_crop in 21 papers lyra/mm_utils.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · c80358dff96c25ce · report

Tasks

MMEVisual Question AnsweringVisual Question Answering (VQA)

5 archive task tags without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Visual Question Answering MM-Vet Lyra-Pro GPT-4 score 71.4 #9 of 231 Archive leaderboard report
Visual Question Answering MM-Vet Lyra-Pro Params 74B #9 of 231 Archive leaderboard report
Visual Question Answering MM-Vet Lyra-Base GPT-4 score 63.5 #27 of 231 Archive leaderboard report
Visual Question Answering MM-Vet Lyra-Base Params 9B #27 of 231 Archive leaderboard report
Visual Question Answering MM-Vet Lyra-Mini GPT-4 score 51.2 #59 of 231 Archive leaderboard report
Visual Question Answering MM-Vet Lyra-Mini Params 3B #59 of 231 Archive leaderboard report
Visual Question Answering (VQA) EgoSchema Lyra-Pro Acc 75.8 #1 of 1 Archive leaderboard report
Visual Question Answering (VQA) MM-Vet Lyra-Pro Acc 71.4 #1 of 1 Archive leaderboard report
Visual Question Answering (VQA) MME Lyra-Pro Acc 2485 #1 of 1 Archive leaderboard report
Visual Question Answering (VQA) MVBench Lyra-Pro Acc 72.3 #1 of 1 Archive leaderboard report
Visual Question Answering (VQA) TextVQA Lyra-Pro Acc 83.5 #1 of 1 Archive leaderboard report
Visual Question Answering (VQA) Video MME Lyra-Pro Acc 69.9 #1 of 1 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

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