GOESB

GOESB — Global Open Edge Speech Benchmark · pronounced “goose-bee”

Leaderboard

Objective, reproducible benchmarks for local (edge) speech AI. Every result here re-verified its hash and signature before being accepted — see why you can trust the numbers. New here? Read the docs.

This is every verified result, filterable and sortable. Looking for a single best recommendation for your constraints instead? Try the homepage →

Lowest energyFastest (lowest RTF)Most accurate (lowest WER)
AnyEvery verified result, batch and streaming mixed together.
AnyBatchStreamingConcurrency

Sorted by most recently submitted · 1 result · click a metric below to sort by it

ProfileRuntime / ModelHardwarePackMetricsSubmittedDetails
whispercpp-medium-en-streamingwhisper-cpp / whisper-medium.enamd-ryzen-7-5700xlibrispeech-en
wer: 0.1104 ratioram_mb: 2148.87 MBcpu_pct: 379.45 %real_time_factor: 7.60 ratioupdate_frequency: 1.04 Hzpartial_stability: 0.8848 ratiofirst_final_latency: 9849.37 msend_of_speech_latency: 7226.41 msfirst_partial_latency: 7612.71 msstreaming_responsiveness: 0.1225 index
2026-08-03T21:50:36Z
details
Profile
whispercpp-medium-en-streaming@1.0.0
Pack
librispeech-en@1.0.0
Repeats
2
Backend
cpu
chunk_ms
1000
threads
4

Constant across this view: chunk_ms 1000 · threads 4

* profile default

26–26 of 26
← PrevPage 2 of 2Next →