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 / ModelHardwarePackconcurrencyMetricsSubmittedDetails
parakeet-tdt-v3-concurrencyparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gblibrispeech-en1 *
ram_mb: 2897.01 MBcpu_pct: 114.56 %throughput: 94.92 audio-s/sreal_time_factor: 0.0123 ratio
2026-08-04T17:35:22Z
details
Profile
parakeet-tdt-v3-concurrency@1.0.0
Pack
librispeech-en@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
concurrency
1
threads
4
parakeet-tdt-v3-concurrencyparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gblibrispeech-en2
ram_mb: 3190.52 MBcpu_pct: 139.04 %throughput: 82.28 audio-s/sreal_time_factor: 0.0285 ratio
2026-08-04T17:36:52Z
details
Profile
parakeet-tdt-v3-concurrency@1.0.0
Pack
librispeech-en@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
concurrency
2 (default 1)
threads
4
parakeet-tdt-v3-concurrencyparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gblibrispeech-en4
ram_mb: 3479.02 MBcpu_pct: 148.38 %throughput: 59.18 audio-s/sreal_time_factor: 0.0793 ratio
2026-08-04T17:38:32Z
details
Profile
parakeet-tdt-v3-concurrency@1.0.0
Pack
librispeech-en@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
concurrency
4 (default 1)
threads
4
parakeet-tdt-v3-de-streamingparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gbfleurs-de—
wer: 0.1080 ratioram_mb: 3241.19 MBcpu_pct: 214.27 %real_time_factor: 0.1273 ratioupdate_frequency: 1.13 Hzpartial_stability: 0.8388 ratiofirst_final_latency: 3904.00 msend_of_speech_latency: 1936.58 msfirst_partial_latency: 1259.60 msstreaming_responsiveness: 1.13 index
2026-08-04T17:08:15Z
details
Profile
parakeet-tdt-v3-de-streaming@1.0.0
Pack
fleurs-de@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
chunk_ms
1000
threads
4
parakeet-tdt-v3-en-streamingparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gblibrispeech-en—
wer: 0.1164 ratioram_mb: 3016.66 MBcpu_pct: 203.40 %real_time_factor: 0.1444 ratioupdate_frequency: 1.01 Hzpartial_stability: 0.7857 ratiofirst_final_latency: 3326.36 msend_of_speech_latency: 890.50 msfirst_partial_latency: 967.55 msstreaming_responsiveness: 0.9541 index
2026-08-04T17:09:14Z
details
Profile
parakeet-tdt-v3-en-streaming@1.0.0
Pack
librispeech-en@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
chunk_ms
1000
threads
4
parakeet-tdt-v3-es-streamingparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gbfleurs-es—
wer: 0.0399 ratioram_mb: 2803.93 MBcpu_pct: 209.43 %real_time_factor: 0.1490 ratioupdate_frequency: 1.08 Hzpartial_stability: 0.7707 ratiofirst_final_latency: 3609.03 msend_of_speech_latency: 1135.56 msfirst_partial_latency: 1074.12 msstreaming_responsiveness: 0.7955 index
2026-08-04T17:10:22Z
details
Profile
parakeet-tdt-v3-es-streaming@1.0.0
Pack
fleurs-es@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
chunk_ms
1000
threads
4
parakeet-tdt-v3-fr-streamingparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gbfleurs-fr—
wer: 0.0674 ratioram_mb: 3045.04 MBcpu_pct: 184.04 %real_time_factor: 0.1608 ratioupdate_frequency: 1.10 Hzpartial_stability: 0.7779 ratiofirst_final_latency: 3182.87 msend_of_speech_latency: 1102.17 msfirst_partial_latency: 1378.54 msstreaming_responsiveness: 0.5703 index
2026-08-04T17:11:26Z
details
Profile
parakeet-tdt-v3-fr-streaming@1.0.0
Pack
fleurs-fr@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
chunk_ms
1000
threads
4
parakeet-tdt-v3-nl-streamingparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gbcommon-voice-nl-elderly—
wer: 0.0950 ratioram_mb: 3153.00 MBcpu_pct: 209.72 %real_time_factor: 0.1141 ratioupdate_frequency: 1.22 Hzpartial_stability: 0.7639 ratiofirst_final_latency: 3415.21 msend_of_speech_latency: 1445.43 msfirst_partial_latency: 1073.56 msstreaming_responsiveness: 1.02 index
2026-08-04T17:13:23Z
details
Profile
parakeet-tdt-v3-nl-streaming@1.0.0
Pack
common-voice-nl-elderly@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
chunk_ms
1000
threads
4
parakeet-tdt-v3-nl-streamingparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gbfleurs-nl—
wer: 0.0657 ratioram_mb: 3373.65 MBcpu_pct: 179.51 %real_time_factor: 0.1832 ratioupdate_frequency: 1.10 Hzpartial_stability: 0.6351 ratiofirst_final_latency: 3285.41 msend_of_speech_latency: 820.23 msfirst_partial_latency: 1078.73 msstreaming_responsiveness: 0.7560 index
2026-08-04T17:14:24Z
details
Profile
parakeet-tdt-v3-nl-streaming@1.0.0
Pack
fleurs-nl@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
chunk_ms
1000
threads
4
parakeet-tdt-v3-pt-streamingparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gbfleurs-pt—
wer: 0.0246 ratioram_mb: 3120.69 MBcpu_pct: 190.62 %real_time_factor: 0.1449 ratioupdate_frequency: 1.08 Hzpartial_stability: 0.7894 ratiofirst_final_latency: 2954.84 msend_of_speech_latency: 1101.30 msfirst_partial_latency: 1111.39 msstreaming_responsiveness: 0.8373 index
2026-08-04T17:15:39Z
details
Profile
parakeet-tdt-v3-pt-streaming@1.0.0
Pack
fleurs-pt@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
chunk_ms
1000
threads
4
parakeet-tdt-v3-de-batchparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gbfleurs-de—
cer: 0.0193 ratiower: 0.0400 ratioram_mb: 2847.10 MBcpu_pct: 314.87 %real_time_factor: 0.0174 ratio
2026-08-04T15:52:03Z
details
Profile
parakeet-tdt-v3-de-batch@1.0.0
Pack
fleurs-de@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
threads
4
parakeet-tdt-v3-en-batchparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gblibrispeech-en—
cer: 0.0081 ratiower: 0.0299 ratioram_mb: 2916.78 MBcpu_pct: 112.39 %real_time_factor: 0.0110 ratio
2026-08-04T15:52:23Z
details
Profile
parakeet-tdt-v3-en-batch@1.0.0
Pack
librispeech-en@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
threads
4
parakeet-tdt-v3-es-batchparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gbfleurs-es—
cer: 0.0052 ratiower: 0.0085 ratioram_mb: 3034.33 MBcpu_pct: 88.42 %real_time_factor: 0.0099 ratio
2026-08-04T15:52:48Z
details
Profile
parakeet-tdt-v3-es-batch@1.0.0
Pack
fleurs-es@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
threads
4
parakeet-tdt-v3-fr-batchparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gbfleurs-fr—
cer: 0.0131 ratiower: 0.0440 ratioram_mb: 3013.72 MBcpu_pct: 118.49 %real_time_factor: 0.0122 ratio
2026-08-04T15:53:13Z
details
Profile
parakeet-tdt-v3-fr-batch@1.0.0
Pack
fleurs-fr@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
threads
4
parakeet-tdt-v3-nl-batchparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gbcommon-voice-nl—
cer: 0.0768 ratiower: 0.1879 ratioram_mb: 2592.62 MBcpu_pct: 121.93 %real_time_factor: 0.0140 ratio
2026-08-04T15:56:46Z
details
Profile
parakeet-tdt-v3-nl-batch@1.0.0
Pack
common-voice-nl@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
threads
4
parakeet-tdt-v3-nl-batchparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gbcommon-voice-nl-elderly—
cer: 0.0107 ratiower: 0.0343 ratioram_mb: 3072.94 MBcpu_pct: 150.91 %real_time_factor: 0.0111 ratio
2026-08-04T15:57:27Z
details
Profile
parakeet-tdt-v3-nl-batch@1.0.0
Pack
common-voice-nl-elderly@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
threads
4
parakeet-tdt-v3-nl-batchparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gbfleurs-nl—
cer: 0.0201 ratiower: 0.0620 ratioram_mb: 3400.97 MBcpu_pct: 114.93 %real_time_factor: 0.0128 ratio
2026-08-04T15:57:49Z
details
Profile
parakeet-tdt-v3-nl-batch@1.0.0
Pack
fleurs-nl@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
threads
4
parakeet-tdt-v3-pt-batchparakeet / parakeet-tdt-0.6b-v3cudanvidia-rtx-4060-ti-16gbfleurs-pt—
cer: 0.0036 ratiower: 0.0137 ratioram_mb: 3214.95 MBcpu_pct: 119.68 %real_time_factor: 0.0092 ratio
2026-08-04T15:58:13Z
details
Profile
parakeet-tdt-v3-pt-batch@1.0.0
Pack
fleurs-pt@1.0.0
Repeats
2
Backend
cuda
Submitted by
taktx#fbc9efc5
threads
4
whisper-base-de-streamingfaster-whisper / whisper-basecudanvidia-rtx-4060-ti-16gbfleurs-de—
wer: 0.1760 ratioram_mb: 792.03 MBcpu_pct: 1180.53 %real_time_factor: 0.0687 ratioupdate_frequency: 1.33 Hzpartial_stability: 0.8083 ratiofirst_final_latency: 3715.51 msend_of_speech_latency: 2476.05 msfirst_partial_latency: 1028.61 msstreaming_responsiveness: 1.06 index
2026-08-03T10:42:34Z
details
Profile
whisper-base-de-streaming@1.0.0
Pack
fleurs-de@1.0.0
Repeats
2
Backend
cuda
beam_size
5
chunk_ms
1000
quantization
int8
threads
4
vad
true
whisper-base-en-streamingfaster-whisper / whisper-basecudanvidia-rtx-4060-ti-16gblibrispeech-en—
wer: 0.1254 ratioram_mb: 770.65 MBcpu_pct: 1331.58 %real_time_factor: 0.0796 ratioupdate_frequency: 1.09 Hzpartial_stability: 0.8334 ratiofirst_final_latency: 3501.64 msend_of_speech_latency: 696.54 msfirst_partial_latency: 825.40 msstreaming_responsiveness: 1.04 index
2026-08-03T10:43:03Z
details
Profile
whisper-base-en-streaming@1.0.0
Pack
librispeech-en@1.0.0
Repeats
2
Backend
cuda
beam_size
5
chunk_ms
1000
quantization
int8
threads
4
vad
true
whisper-base-es-streamingfaster-whisper / whisper-basecudanvidia-rtx-4060-ti-16gbfleurs-es—
wer: 0.1282 ratioram_mb: 774.35 MBcpu_pct: 1234.56 %real_time_factor: 0.0810 ratioupdate_frequency: 1.17 Hzpartial_stability: 0.7659 ratiofirst_final_latency: 3739.12 msend_of_speech_latency: 1499.18 msfirst_partial_latency: 1182.56 msstreaming_responsiveness: 0.7489 index
2026-08-03T10:43:51Z
details
Profile
whisper-base-es-streaming@1.0.0
Pack
fleurs-es@1.0.0
Repeats
2
Backend
cuda
beam_size
5
chunk_ms
1000
quantization
int8
threads
4
vad
true
whisper-base-fr-streamingfaster-whisper / whisper-basecudanvidia-rtx-4060-ti-16gbfleurs-fr—
wer: 0.4633 ratioram_mb: 759.37 MBcpu_pct: 1119.88 %real_time_factor: 0.0915 ratioupdate_frequency: 1.22 Hzpartial_stability: 0.6392 ratiofirst_final_latency: 3456.56 msend_of_speech_latency: 1426.09 msfirst_partial_latency: 1061.01 msstreaming_responsiveness: 0.7424 index
2026-08-03T10:44:28Z
details
Profile
whisper-base-fr-streaming@1.0.0
Pack
fleurs-fr@1.0.0
Repeats
2
Backend
cuda
beam_size
5
chunk_ms
1000
quantization
int8
threads
4
vad
true
whisper-base-nl-streamingfaster-whisper / whisper-basecudanvidia-rtx-4060-ti-16gbfleurs-nl—
wer: 0.3504 ratioram_mb: 775.83 MBcpu_pct: 1076.00 %real_time_factor: 0.1032 ratioupdate_frequency: 1.24 Hzpartial_stability: 0.5498 ratiofirst_final_latency: 3778.78 msend_of_speech_latency: 1148.04 msfirst_partial_latency: 890.19 msstreaming_responsiveness: 0.9098 index
2026-08-03T10:45:00Z
details
Profile
whisper-base-nl-streaming@1.0.0
Pack
fleurs-nl@1.0.0
Repeats
2
Backend
cuda
beam_size
5
chunk_ms
1000
quantization
int8
threads
4
vad
true
whisper-base-pt-streamingfaster-whisper / whisper-basecudanvidia-rtx-4060-ti-16gbfleurs-pt—
wer: 0.2541 ratioram_mb: 774.98 MBcpu_pct: 1257.24 %real_time_factor: 0.0867 ratioupdate_frequency: 1.16 Hzpartial_stability: 0.7131 ratiofirst_final_latency: 3682.94 msend_of_speech_latency: 1447.39 msfirst_partial_latency: 1022.92 msstreaming_responsiveness: 0.8450 index
2026-08-03T10:45:46Z
details
Profile
whisper-base-pt-streaming@1.0.0
Pack
fleurs-pt@1.0.0
Repeats
2
Backend
cuda
beam_size
5
chunk_ms
1000
quantization
int8
threads
4
vad
true
whisper-large-v3-de-streamingfaster-whisper / whisper-large-v3cudanvidia-rtx-4060-ti-16gbfleurs-de—
wer: 0.1160 ratioram_mb: 3673.01 MBcpu_pct: 377.09 %real_time_factor: 0.3290 ratioupdate_frequency: 1.22 Hzpartial_stability: 0.7815 ratiofirst_final_latency: 3168.54 msend_of_speech_latency: 2857.09 msfirst_partial_latency: 1189.32 msstreaming_responsiveness: 0.8212 index
2026-08-03T10:48:18Z
details
Profile
whisper-large-v3-de-streaming@1.0.0
Pack
fleurs-de@1.0.0
Repeats
2
Backend
cuda
beam_size
5
chunk_ms
1000
quantization
int8
threads
4
vad
true

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

* profile default

1–25 of 48
← PrevPage 1 of 2Next →