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

ProfileRuntime / ModelHardwarePackMetricsSubmittedDetails
whispercpp-small-nl-streamingwhisper-cpp / whisper-smallmetalapple-m1-profleurs-nl
wer: 0.2993 ratioram_mb: 887.98 MBcpu_pct: 21.73 %real_time_factor: 0.2904 ratioupdate_frequency: 1.13 Hzpartial_stability: 0.7002 ratiofirst_final_latency: 3648.23 msend_of_speech_latency: 1135.16 msfirst_partial_latency: 1103.56 msstreaming_responsiveness: 0.7531 index
2026-08-03T15:13:53Z
details
Profile
whispercpp-small-nl-streaming@1.0.0
Pack
fleurs-nl@1.0.0
Repeats
2
Backend
metal
chunk_ms
1000
threads
4
whispercpp-small-pt-streamingwhisper-cpp / whisper-smallmetalapple-m1-profleurs-pt
wer: 0.1093 ratioram_mb: 894.16 MBcpu_pct: 20.46 %real_time_factor: 0.2553 ratioupdate_frequency: 1.00 Hzpartial_stability: 0.8991 ratiofirst_final_latency: 4250.31 msend_of_speech_latency: 1342.91 msfirst_partial_latency: 1177.90 msstreaming_responsiveness: 0.7621 index
2026-08-03T15:15:32Z
details
Profile
whispercpp-small-pt-streaming@1.0.0
Pack
fleurs-pt@1.0.0
Repeats
2
Backend
metal
chunk_ms
1000
threads
4
whispercpp-tiny-de-streamingwhisper-cpp / whisper-tinymetalapple-m1-profleurs-de
wer: 0.3640 ratioram_mb: 314 MBcpu_pct: 55.72 %real_time_factor: 0.0744 ratioupdate_frequency: 1.00 Hzpartial_stability: 0.8314 ratiofirst_final_latency: 4856.19 msend_of_speech_latency: 2061.65 msfirst_partial_latency: 977.94 msstreaming_responsiveness: 0.8048 index
2026-08-03T15:15:58Z
details
Profile
whispercpp-tiny-de-streaming@1.0.0
Pack
fleurs-de@1.0.0
Repeats
2
Backend
metal
chunk_ms
1000
threads
4
whispercpp-tiny-en-streamingwhisper-cpp / whisper-tiny.enmetalapple-m1-prolibrispeech-en
wer: 0.1522 ratioram_mb: 303.29 MBcpu_pct: 42.24 %real_time_factor: 0.0588 ratioupdate_frequency: 1.04 Hzpartial_stability: 0.8716 ratiofirst_final_latency: 3120.55 msend_of_speech_latency: 436.61 msfirst_partial_latency: 897.50 msstreaming_responsiveness: 0.8867 index
2026-08-03T15:16:17Z
details
Profile
whispercpp-tiny-en-streaming@1.0.0
Pack
librispeech-en@1.0.0
Repeats
2
Backend
metal
chunk_ms
1000
whispercpp-tiny-es-streamingwhisper-cpp / whisper-tinymetalapple-m1-profleurs-es
wer: 0.2336 ratioram_mb: 303.17 MBcpu_pct: 48.70 %real_time_factor: 0.0794 ratioupdate_frequency: 1.05 Hzpartial_stability: 0.8391 ratiofirst_final_latency: 4503.03 msend_of_speech_latency: 1276.07 msfirst_partial_latency: 895.69 msstreaming_responsiveness: 1.06 index
2026-08-03T15:16:45Z
details
Profile
whispercpp-tiny-es-streaming@1.0.0
Pack
fleurs-es@1.0.0
Repeats
2
Backend
metal
chunk_ms
1000
threads
4
whispercpp-tiny-fr-streamingwhisper-cpp / whisper-tinymetalapple-m1-profleurs-fr
wer: 0.5836 ratioram_mb: 303.83 MBcpu_pct: 52.56 %real_time_factor: 0.0978 ratioupdate_frequency: 1.06 Hzpartial_stability: 0.6572 ratiofirst_final_latency: 4519.86 msend_of_speech_latency: 998.14 msfirst_partial_latency: 852.20 msstreaming_responsiveness: 0.6977 index
2026-08-03T15:17:15Z
details
Profile
whispercpp-tiny-fr-streaming@1.0.0
Pack
fleurs-fr@1.0.0
Repeats
2
Backend
metal
chunk_ms
1000
threads
4
whispercpp-tiny-nl-streamingwhisper-cpp / whisper-tinymetalapple-m1-profleurs-nl
wer: 0.6679 ratioram_mb: 306.97 MBcpu_pct: 57.50 %real_time_factor: 0.1031 ratioupdate_frequency: 1.16 Hzpartial_stability: 0.6170 ratiofirst_final_latency: 4195.81 msend_of_speech_latency: 921.43 msfirst_partial_latency: 845.99 msstreaming_responsiveness: 0.8358 index
2026-08-03T15:17:40Z
details
Profile
whispercpp-tiny-nl-streaming@1.0.0
Pack
fleurs-nl@1.0.0
Repeats
2
Backend
metal
chunk_ms
1000
threads
4
whispercpp-tiny-pt-streamingwhisper-cpp / whisper-tinymetalapple-m1-profleurs-pt
wer: 0.2951 ratioram_mb: 313.28 MBcpu_pct: 57.05 %real_time_factor: 0.1043 ratioupdate_frequency: 1.07 Hzpartial_stability: 0.8262 ratiofirst_final_latency: 4312.57 msend_of_speech_latency: 1270.45 msfirst_partial_latency: 944.50 msstreaming_responsiveness: 0.8474 index
2026-08-03T15:18:21Z
details
Profile
whispercpp-tiny-pt-streaming@1.0.0
Pack
fleurs-pt@1.0.0
Repeats
2
Backend
metal
chunk_ms
1000
threads
4
whisper-base-de-batchfaster-whisper / whisper-basecudanvidia-rtx-5090fleurs-de
cer: 0.0320 ratiower: 0.0920 ratioram_mb: 857.43 MBcpu_pct: 567.65 %temperature_c: 52.25 °Creal_time_factor: 0.0176 ratio
2026-08-01T13:56:10Z
details
Profile
whisper-base-de-batch@1.1.0
Pack
fleurs-de@1.0.0
Repeats
2
Backend
cuda
beam_size
5
quantization
int8
threads
4
vad
true
whisper-base-en-batchfaster-whisper / whisper-basecudanvidia-rtx-5090librispeech-en
cer: 0.0433 ratiower: 0.1015 ratioram_mb: 899.47 MBcpu_pct: 558.11 %temperature_c: 54.63 °Creal_time_factor: 0.0205 ratio
2026-08-01T13:56:18Z
details
Profile
whisper-base-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-base-es-batchfaster-whisper / whisper-basecudanvidia-rtx-5090fleurs-es
cer: 0.0164 ratiower: 0.0684 ratioram_mb: 859.77 MBcpu_pct: 518.70 %temperature_c: 55.06 °Creal_time_factor: 0.0183 ratio
2026-08-01T13:56:26Z
details
Profile
whisper-base-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-base-fr-batchfaster-whisper / whisper-basecudanvidia-rtx-5090fleurs-fr
cer: 0.0692 ratiower: 0.2082 ratioram_mb: 873.71 MBcpu_pct: 514.63 %temperature_c: 54.38 °Creal_time_factor: 0.0217 ratio
2026-08-01T13:56:34Z
details
Profile
whisper-base-fr-batch@1.1.0
Pack
fleurs-fr@1.0.0
Repeats
2
Backend
cuda
beam_size
5
quantization
int8
threads
4
vad
true
whisper-base-nl-batchfaster-whisper / whisper-basecudanvidia-rtx-5090fleurs-nl
cer: 0.1019 ratiower: 0.3066 ratioram_mb: 857.14 MBcpu_pct: 538.30 %temperature_c: 52.63 °Creal_time_factor: 0.0265 ratio
2026-08-01T13:56:43Z
details
Profile
whisper-base-nl-batch@1.1.0
Pack
fleurs-nl@1.0.0
Repeats
2
Backend
cuda
beam_size
5
quantization
int8
threads
4
vad
true
whisper-base-pt-batchfaster-whisper / whisper-basecudanvidia-rtx-5090fleurs-pt
cer: 0.0164 ratiower: 0.0519 ratioram_mb: 862.34 MBcpu_pct: 501.57 %temperature_c: 52.19 °Creal_time_factor: 0.0189 ratio
2026-08-01T13:56:52Z
details
Profile
whisper-base-pt-batch@1.1.0
Pack
fleurs-pt@1.0.0
Repeats
2
Backend
cuda
beam_size
5
quantization
int8
threads
4
vad
true
whisper-large-v3-de-batchfaster-whisper / whisper-large-v3cudanvidia-rtx-5090fleurs-de
cer: 0.0054 ratiower: 0.0160 ratioram_mb: 3735.71 MBcpu_pct: 228.08 %temperature_c: 53.69 °Creal_time_factor: 0.0265 ratio
2026-08-01T13:57:39Z
details
Profile
whisper-large-v3-de-batch@1.1.0
Pack
fleurs-de@1.0.0
Repeats
2
Backend
cuda
beam_size
5
quantization
int8
threads
4
vad
true
whisper-large-v3-en-batchfaster-whisper / whisper-large-v3cudanvidia-rtx-5090librispeech-en
cer: 0.0276 ratiower: 0.0597 ratioram_mb: 3545.36 MBcpu_pct: 377.26 %temperature_c: 53.75 °Creal_time_factor: 0.0329 ratio
2026-08-01T13:57:57Z
details
Profile
whisper-large-v3-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-large-v3-es-batchfaster-whisper / whisper-large-v3cudanvidia-rtx-5090fleurs-es
cer: 0.0052 ratiower: 0.0085 ratioram_mb: 3595.43 MBcpu_pct: 359.23 %temperature_c: 53.13 °Creal_time_factor: 0.0322 ratio
2026-08-01T13:58:17Z
details
Profile
whisper-large-v3-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-large-v3-fr-batchfaster-whisper / whisper-large-v3cudanvidia-rtx-5090fleurs-fr
cer: 0.0127 ratiower: 0.0440 ratioram_mb: 3531.94 MBcpu_pct: 329.03 %temperature_c: 54.50 °Creal_time_factor: 0.0370 ratio
2026-08-01T13:58:37Z
details
Profile
whisper-large-v3-fr-batch@1.1.0
Pack
fleurs-fr@1.0.0
Repeats
2
Backend
cuda
beam_size
5
quantization
int8
threads
4
vad
true
whisper-large-v3-nl-batchfaster-whisper / whisper-large-v3cudanvidia-rtx-5090fleurs-nl
cer: 0.0095 ratiower: 0.0401 ratioram_mb: 3550.39 MBcpu_pct: 353.01 %temperature_c: 52.38 °Creal_time_factor: 0.0442 ratio
2026-08-01T13:58:57Z
details
Profile
whisper-large-v3-nl-batch@1.1.0
Pack
fleurs-nl@1.0.0
Repeats
2
Backend
cuda
beam_size
5
quantization
int8
threads
4
vad
true
whisper-large-v3-pt-batchfaster-whisper / whisper-large-v3cudanvidia-rtx-5090fleurs-pt
cer: 0.0018 ratiower: 0.0082 ratioram_mb: 3497.17 MBcpu_pct: 370.52 %temperature_c: 53.56 °Creal_time_factor: 0.0296 ratio
2026-08-01T13:59:17Z
details
Profile
whisper-large-v3-pt-batch@1.1.0
Pack
fleurs-pt@1.0.0
Repeats
2
Backend
cuda
beam_size
5
quantization
int8
threads
4
vad
true
whisper-medium-de-batchfaster-whisper / whisper-mediumcudanvidia-rtx-5090fleurs-de
cer: 0.0079 ratiower: 0.0360 ratioram_mb: 2228.55 MBcpu_pct: 299.42 %temperature_c: 52.38 °Creal_time_factor: 0.0262 ratio
2026-08-01T13:59:46Z
details
Profile
whisper-medium-de-batch@1.1.0
Pack
fleurs-de@1.0.0
Repeats
2
Backend
cuda
beam_size
5
quantization
int8
threads
4
vad
true
whisper-medium-en-batchfaster-whisper / whisper-mediumcudanvidia-rtx-5090librispeech-en
cer: 0.0352 ratiower: 0.0836 ratioram_mb: 2002.75 MBcpu_pct: 408.37 %temperature_c: 51.75 °Creal_time_factor: 0.0292 ratio
2026-08-01T14:00:00Z
details
Profile
whisper-medium-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-medium-es-batchfaster-whisper / whisper-mediumcudanvidia-rtx-5090fleurs-es
cer: 0.0066 ratiower: 0.0171 ratioram_mb: 2033.13 MBcpu_pct: 406.39 %temperature_c: 51.06 °Creal_time_factor: 0.0281 ratio
2026-08-01T14:00:15Z
details
Profile
whisper-medium-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-medium-fr-batchfaster-whisper / whisper-mediumcudanvidia-rtx-5090fleurs-fr
cer: 0.0348 ratiower: 0.0909 ratioram_mb: 2042.20 MBcpu_pct: 387.00 %temperature_c: 51.69 °Creal_time_factor: 0.0338 ratio
2026-08-01T14:00:30Z
details
Profile
whisper-medium-fr-batch@1.1.0
Pack
fleurs-fr@1.0.0
Repeats
2
Backend
cuda
beam_size
5
quantization
int8
threads
4
vad
true
whisper-medium-nl-batchfaster-whisper / whisper-mediumcudanvidia-rtx-5090fleurs-nl
cer: 0.0162 ratiower: 0.0511 ratioram_mb: 1999.21 MBcpu_pct: 383.46 %temperature_c: 52.38 °Creal_time_factor: 0.0391 ratio
2026-08-01T14:00:45Z
details
Profile
whisper-medium-nl-batch@1.1.0
Pack
fleurs-nl@1.0.0
Repeats
2
Backend
cuda
beam_size
5
quantization
int8
threads
4
vad
true

Constant across this view: beam_size 5 · chunk_ms 1000 · quantization int8 · threads 4 · vad true

* profile default

101–125 of 556
← PrevPage 5 of 23Next →