SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2505.18614 · EMNLP · 2025

MAVL: A Multilingual Audio-Video Lyrics Dataset for Animated Song Translation

Youngjae Yu, Youngmin Kim, Sunghyun Lee, Woohyun Cho

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 19 functions out of this paper's own repositories and ran 13 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
jianfch/stable-ts canonical 13 of 19
FunctionStatusWhere it lives
as_vanilla Ran jianfch/stable-ts/stable_whisper/whisper_compatibility.py
code served (permissive licence) · get_code("edd87e3d5219c886")
dfnet_audio Ran jianfch/stable-ts/stable_whisper/audio/dfnet.py
code served (permissive licence) · get_code("a2906ebf1c03e2b2")
encode_video_comparison Ran jianfch/stable-ts/stable_whisper/video_output.py
code served (permissive licence) · get_code("f0b041c207e1f563")
finalize_text Ran jianfch/stable-ts/stable_whisper/text_output.py
code served (permissive licence) · get_code("88205be653cfa848")
get_func_parameters Ran jianfch/stable-ts/stable_whisper/utils.py
code served (permissive licence) · get_code("2f71dfac10f2811e")
get_val Ran jianfch/stable-ts/stable_whisper/default.py
code served (permissive licence) · get_code("8c364004420fb08f")
has_key Ran jianfch/stable-ts/stable_whisper/default.py
code served (permissive licence) · get_code("2cb0d3956bd3ca79")
isolate_useful_options Ran jianfch/stable-ts/stable_whisper/utils.py
code served (permissive licence) · get_code("9595a2a36e654ced")
sec2hhmmss Ran jianfch/stable-ts/stable_whisper/text_output.py
code served (permissive licence) · get_code("1385e684ee00e274")
sec2milliseconds Ran jianfch/stable-ts/stable_whisper/text_output.py
code served (permissive licence) · get_code("940741f0689b881e")
set_get_val Ran jianfch/stable-ts/stable_whisper/default.py
code served (permissive licence) · get_code("6e5e6a677cc59eed")
split_word_tokens Ran jianfch/stable-ts/stable_whisper/timing.py
code served (permissive licence) · get_code("fa9a731acd304668")
str_to_valid_type Ran jianfch/stable-ts/stable_whisper/utils.py
code served (permissive licence) · get_code("0c886f090ffcc1ab")
find_alignment_stable Not yet run jianfch/stable-ts/stable_whisper/timing.py
code served (permissive licence) · get_code("7b13c54749fcc4a8")
get_alignment_tokenizer Not yet run jianfch/stable-ts/stable_whisper/alignment.py
code served (permissive licence) · get_code("50582ac3612eaa85")
get_tokenizer Not yet run jianfch/stable-ts/stable_whisper/whisper_compatibility.py
code served (permissive licence) · get_code("8eb94fa43ead16a2")
get_valid_language Not yet run jianfch/stable-ts/stable_whisper/whisper_compatibility.py
code served (permissive licence) · get_code("52e3fba663460cae")
load_dfnet_model Not yet run jianfch/stable-ts/stable_whisper/audio/dfnet.py
code served (permissive licence) · get_code("4a82dc494d002e7f")
pop_empty_alignment Not yet run jianfch/stable-ts/stable_whisper/timing.py
code served (permissive licence) · get_code("4d0397204f102619")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Lyrics translation requires both accurate semantic transfer and preservation of musical rhythm, syllabic structure, and poetic style. In animated musicals, the challenge intensifies due to alignment with visual and auditory cues. We introduce Multilingual Audio-Video Lyrics Benchmark for Animated Song Translation (MAVL), the first multilingual, multimodal benchmark for singable lyrics translation. By integrating text, audio, and video, MAVL enables richer and more expressive translations than textonly approaches. Building on this, we propose Syllable-Constrained Audio-Video LLM with Chain-of-Thought (SylAVL-CoT), which leverages audio-video cues and enforces syllabic constraints to produce natural-sounding lyrics. Experimental results demonstrate that SylAVL-CoT significantly outperforms textbased models in singability and contextual accuracy, emphasizing the value of multimodal, multilingual approaches for lyrics translation.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2505.18614")
get_code_for_paper("2505.18614")
have("2505.18614")

Connect an agent — have() is free.