Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Audio Flamingo 3: Advancing Audio Intelligence with Fully Open Large Audio Language Models

About

We present Audio Flamingo 3 (AF3), a fully open state-of-the-art (SOTA) large audio-language model that advances reasoning and understanding across speech, sound, and music. AF3 introduces: (i) AF-Whisper, a unified audio encoder trained using a novel strategy for joint representation learning across all 3 modalities of speech, sound, and music; (ii) flexible, on-demand thinking, allowing the model to do chain-of-thought-type reasoning before answering; (iii) multi-turn, multi-audio chat; (iv) long audio understanding and reasoning (including speech) up to 10 minutes; and (v) voice-to-voice interaction. To enable these capabilities, we propose several large-scale training datasets curated using novel strategies, including AudioSkills-XL, LongAudio-XL, AF-Think, and AF-Chat, and train AF3 with a novel five-stage curriculum-based training strategy. Trained on only open-source audio data, AF3 achieves new SOTA results on over 20+ (long) audio understanding and reasoning benchmarks, surpassing both open-weight and closed-source models trained on much larger datasets.

Arushi Goel, Sreyan Ghosh, Jaehyeon Kim, Sonal Kumar, Zhifeng Kong, Sang-gil Lee, Chao-Han Huck Yang, Ramani Duraiswami, Dinesh Manocha, Rafael Valle, Bryan Catanzaro• 2025

Related benchmarks

TaskDatasetResultRank
Automatic Speech RecognitionLibriSpeech clean (test)
WER1.57
1156
Automatic Speech RecognitionLibriSpeech (test-other)
WER3.13
1151
Audio CaptioningAudioCaps (test)
CIDEr0.79
140
Automatic Speech RecognitionLibriSpeech Other
WER3.13
96
Multimodal UnderstandingMMMU
MMMU Score72.42
59
Audio UnderstandingMMAU v05.15.25 (test)
Sound Score79.58
53
Audio CaptioningAudioCaps
CIDEr70
49
Audio UnderstandingMMAU v05.15.25 (test-mini)
Sound Score79.58
44
Audio ReasoningMMAR (test)
Sound Score56.97
38
Audio UnderstandingMMSU
Perception Score45.63
32
Showing 10 of 106 rows
...

Other info

Follow for update