Papers › Whisper-Flamingo: Integrating Visual Features into Whisper for Audio-Visual Speech...

Whisper-Flamingo: Integrating Visual Features into Whisper for Audio-Visual Speech Recognition and Translation

14 Jun 2024arXiv:2406.10082archive 2025-07-28

Andrew Rouditchenko, Yuan Gong, Samuel Thomas, Leonid Karlinsky, Hilde Kuehne, Rogerio Feris, James Glass

Audio-Visual Speech Recognition (AVSR) uses lip-based video to improve performance in noise. Since videos are harder to obtain than audio, the video training data of AVSR models is usually limited to a few thousand hours. In contrast, speech models such as Whisper are trained with hundreds of thousands of hours of data, and thus learn a better speech-to-text decoder. The huge training data difference motivates us to adapt Whisper to handle video inputs. Inspired by Flamingo which injects visual features into language models, we propose Whisper-Flamingo which integrates visual features into the Whisper speech recognition and translation model with gated cross attention. Our models achieve state-of-the-art ASR WER (0.68%) and AVSR WER (0.76%) on LRS3, and state-of-the-art ASR WER (1.3%) and AVSR WER (1.4%) on LRS2. Audio-visual Whisper-Flamingo outperforms audio-only Whisper on English speech recognition and En-X translation for 6 languages in noisy conditions. Moreover, Whisper-Flamingo is versatile and conducts all of these tasks using one set of parameters, while prior methods are trained separately on each language.

PaperPDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2406.10082")

Code

Syntology Ran 5 of 18 code samples harvested from 1 repository linked to this paper; 13 have no recorded run. Of those that ran: 2 ran · honoured contract; 1 ran · our draft was wrong; 2 ran with no contract checked.

By repository: official repository: 18 samples from 1 repository, 5 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

roudimit/whisper-flamingo officialmentioned in papermentioned on GitHubpytorchNOASSERTION 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

18 samples harvested; 5 ran; 2 honoured the contract we drafted; 13 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · honoured contract
1ran · our draft was wrong
2ran
13unverified

Licence: 18 of the 18 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 roudimit/whisper-flamingo. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

conv3x3 roudimit/whisper-flamingo/whisper/resnet.py official repository ran · our draft was wrong no licence file found · pointer only · 583f9780bdd00a45 · report
downsample_basic_block roudimit/whisper-flamingo/whisper/resnet.py official repository ran no licence file found · pointer only · d50e3a8f0b0f9a21 · report
downsample_basic_block_v2 roudimit/whisper-flamingo/whisper/resnet.py official repository ran no licence file found · pointer only · 546a770034e200d1 · report
exact_div roudimit/whisper-flamingo/whisper/utils.py official repository ran · honoured contract fingerprinted no licence file found · pointer only · 6e0808fc4828513a · report
sinusoids roudimit/whisper-flamingo/whisper/model.py official repository ran · honoured contract fingerprinted no licence file found · pointer only · e529c9641c178fe0 · report
backtrace roudimit/whisper-flamingo/whisper/timing.py official repository unverified licence not identified · pointer only · 41c654294ba9ec56 · report
dtw_cpu roudimit/whisper-flamingo/whisper/timing.py official repository unverified licence not identified · pointer only · 76330ed965e60af5 · report
get_encoding roudimit/whisper-flamingo/whisper/tokenizer.py official repository unverified licence not identified · pointer only · 1549b2f905088001 · report
get_tokenizer roudimit/whisper-flamingo/whisper/tokenizer.py official repository unverified licence not identified · pointer only · 7929dec299b10283 · report
load_audio roudimit/whisper-flamingo/whisper/audio.py official repository unverified no licence file found · pointer only · 80abc0b281d75575 · report
median_filter roudimit/whisper-flamingo/whisper/timing.py official repository unverified no licence file found · pointer only · 6e70207ad77f0ae6 · report
median_filter_cuda roudimit/whisper-flamingo/whisper/triton_ops.py official repository unverified licence not identified · pointer only · 27325bde549014b8 · report
median_kernel roudimit/whisper-flamingo/whisper/triton_ops.py official repository unverified licence not identified · pointer only · b946e8e3af390ae7 · report
mel_filters roudimit/whisper-flamingo/whisper/audio.py official repository unverified no licence file found · pointer only · 12c6c409a1e359c3 · report
optional_int roudimit/whisper-flamingo/whisper/utils.py official repository unverified no licence file found · pointer only · e56996b20b9c19cb · report
pad_or_trim roudimit/whisper-flamingo/whisper/audio.py official repository unverified no licence file found · pointer only · 9f13de1c23e4003d · report
str2bool roudimit/whisper-flamingo/whisper/utils.py official repository unverified no licence file found · pointer only · 176b6ba6c2c2940c · report
transcribe roudimit/whisper-flamingo/whisper/transcribe.py official repository unverified licence not identified · pointer only · 7bebc0cb7f2ab5b0 · report

Tasks

Audio-Visual Speech RecognitionAutomatic Speech Recognition (ASR)DecoderSpeech RecognitionSpeech-to-TextTranslationVisual Speech Recognitionspeech-recognition

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Audio-Visual Speech Recognition LRS2 Whisper-Flamingo Test WER 1.4 #1 of 8 Archive leaderboard report
Audio-Visual Speech Recognition LRS3-TED Whisper-Flamingo Word Error Rate (WER) 0.76 #2 of 12 Archive leaderboard report
Automatic Speech Recognition (ASR) LRS2 Whisper Test WER 1.3 #1 of 9 Archive leaderboard report
Speech Recognition LRS3-TED Whisper Word Error Rate (WER) 0.68 #1 of 4 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