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 · 3 results · click a metric below to sort by it

ProfileRuntime / ModelHardwarePackMetricsSubmittedDetails
whisper-tiny-en-batchfaster-whisper / whisper-tinycudanvidia-rtx-4090librispeech-en
cer: 0.0520 ratiower: 0.1194 ratioram_mb: 732.55 MBcpu_pct: 1248.97 %temperature_c: 51 °Creal_time_factor: 0.0206 ratio
2026-07-31T08:39:23Z
details
Profile
whisper-tiny-en-batch@1.1.0
Pack
librispeech-en@1.0.0
Repeats
2
Backend
cuda
beam_size
5
quantization
int8
threads
4
vad
true
whisper-tiny-es-batchfaster-whisper / whisper-tinycudanvidia-rtx-4090fleurs-es
cer: 0.0263 ratiower: 0.0940 ratioram_mb: 705.13 MBcpu_pct: 1202.67 %temperature_c: 48 °Creal_time_factor: 0.0198 ratio
2026-07-31T08:39:33Z
details
Profile
whisper-tiny-es-batch@1.1.0
Pack
fleurs-es@1.0.0
Repeats
2
Backend
cuda
beam_size
5
quantization
int8
threads
4
vad
true
whisper-tiny-pt-batchfaster-whisper / whisper-tinycudanvidia-rtx-4090fleurs-pt
cer: 0.0286 ratiower: 0.1148 ratioram_mb: 718.45 MBcpu_pct: 1140.89 %temperature_c: 48 °Creal_time_factor: 0.0192 ratio
2026-07-31T08:40:02Z
details
Profile
whisper-tiny-pt-batch@1.1.0
Pack
fleurs-pt@1.0.0
Repeats
2
Backend
cuda
beam_size
5
quantization
int8
threads
4
vad
true

Constant across this view: beam_size 5 · quantization int8 · threads 4 · vad true

* profile default

26–28 of 28
← PrevPage 2 of 2Next →