Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Speech-to-Text Question-Answering on LlamaQ, TriviaQA, WebQ, and OBQA Average
Loading...
57.6
Accuracy
DIFFUSPEECH
12.048
23.874
35.7
47.526
Jan 30, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
DIFFUSPEECH
Type=Diff., S→S capabi...
2026.01
57.6
MinMo
Type=AR, S→S capabilit...
2026.01
54
Qwen2-Audio
Type=AR, S→S capabilit...
2026.01
46.2
Moshi
Type=AR, S→S capabilit...
2026.01
46
Phi-4-Multimodal
Type=AR, S→S capabilit...
2026.01
43.9
Llama-Omni2
Type=AR, S→S capabilit...
2026.01
43.5
DiFFA
Type=Diff., S→S capabi...
2026.01
43.3
SpiritLM
Type=AR, S→S capabilit...
2026.01
20.2
SpeechGPT
Type=AR, S→S capabilit...
2026.01
13.8
Feedback
Search any
task
Search any
task