GAIK SKN dialect test
Models transcribe 600 Finnish dialect interview segments from ten municipalities. The reference uses standard Finnish word forms.
GAIK reports 19 models across 21 scored configurations. It omits the SKN result for the version trained on every municipality.
- Released
- October 2026
- Kind
- Speech recognition and translation
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Language
- Finnish
- Interaction
- Single turn
- Items from
- Real work and use, Another benchmark
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Audio and speech
- Strongest model tested
-
Gemini 3.5 Flash #44 on the capability index
- Last model added
- May 2026
- Scores refreshed
- November 20th, 2026
Ours: main model (parl) tops the leaderboard with 19.4% Gemini 3.5 Flash leads with 34.6% among models released by May 2026
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Gemini 3.5 Flash, at 34.6%. This view shows only models we track, each at its best.
The newest model with a score here came out in May 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1 Ours: main model (parl) 19.4%
- 2 Ours: la-murre version 19.7%
- 3 Ours: first model (v1.0) 21.1%
- 4 Ours: timestamps version 21.4%
- 5 Ours: round 1 22.2%
- 6
Gemini 3.5 Flash API, asked for standard Finnish 34.6%
- 7 Whisper large-v3 40.4%
- 8 Google Chirp 3 (API) API 41.3%
- 9 Whisper large-v3-turbo 42.5%
- 10 ElevenLabs Scribe v2 (API) API 44.9%
- 11 NVIDIA Canary 1B v2 45.6%
- 12 RASMUS Finnish-ASR-Canary-v2 + KenLM 47.6%
- 13 NVIDIA Nemotron 3.5 Transcribe 47.6%
- 14 Orukeet (Parakeet v3 fine-tune) 48.5%
- 15 RASMUS Finnish-ASR-Canary-v2, greedy 50.1%
- 16 NVIDIA Parakeet TDT 0.6B v3 52.5%
- 17 Finnish-NLP whisper-large-finnish-v3 56.6%
- 18 OpenAI gpt-4o-transcribe (API, via Azure) API, via Azure 62.5%
- 19
Qwen3-ASR-1.7B 66.7%
- 20 Finnish-NLP wav2vec2 XLS-R 1B (no LM) 69.9%
Source: Official GAIK SKN dialect test leaderboard
- 1
Gemini 3.5 Flash API, asked for standard Finnish 34.6%
- 2 Whisper large-v3 40.4%
- 3 Whisper large-v3-turbo 42.5%
- 4
Qwen3-ASR-1.7B 66.7%
Source: Official GAIK SKN dialect test leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Ours: main model (parl) | 19.4% | Official leaderboard | ||
| 2 | Ours: la-murre version | 19.7% | Official leaderboard | ||
| 3 | Ours: first model (v1.0) | 21.1% | Official leaderboard | ||
| 4 | Ours: timestamps version | 21.4% | Official leaderboard | ||
| 5 | Ours: round 1 | 22.2% | Official leaderboard | ||
| 6 | | 34.6% | API, asked for standard Finnish | May 2026 | Official leaderboard |
| 7 | Whisper large-v3 | 40.4% | Nov 2023 | Official leaderboard | |
| 8 | Google Chirp 3 (API) | 41.3% | API | Official leaderboard | |
| 9 | Whisper large-v3-turbo | 42.5% | Oct 2024 | Official leaderboard | |
| 10 | ElevenLabs Scribe v2 (API) | 44.9% | API | Official leaderboard | |
| 11 | NVIDIA Canary 1B v2 | 45.6% | Official leaderboard | ||
| 12 | RASMUS Finnish-ASR-Canary-v2 + KenLM | 47.6% | Official leaderboard | ||
| 13 | NVIDIA Nemotron 3.5 Transcribe | 47.6% | Official leaderboard | ||
| 14 | Orukeet (Parakeet v3 fine-tune) | 48.5% | Official leaderboard | ||
| 15 | RASMUS Finnish-ASR-Canary-v2, greedy | 50.1% | Official leaderboard | ||
| 16 | NVIDIA Parakeet TDT 0.6B v3 | 52.5% | Official leaderboard | ||
| 17 | Finnish-NLP whisper-large-finnish-v3 | 56.6% | Official leaderboard | ||
| 18 | OpenAI gpt-4o-transcribe (API, via Azure) | 62.5% | API, via Azure | Official leaderboard | |
| 19 | | 66.7% | Jan 2026 | Official leaderboard | |
| 20 | Finnish-NLP wav2vec2 XLS-R 1B (no LM) | 69.9% | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 34.6% | API, asked for standard Finnish | May 2026 | Official leaderboard |
| 2 | Whisper large-v3 | 40.4% | Nov 2023 | Official leaderboard | |
| 3 | Whisper large-v3-turbo | 42.5% | Oct 2024 | Official leaderboard | |
| 4 | | 66.7% | Jan 2026 | Official leaderboard |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official GAIK SKN dialect test leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official GAIK SKN dialect test leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Audio and video
- Video-MMMU In use 300 expert-level lecture videos with questions testing perception, comprehension and adaptation of knowledge. Accuracy.
Qwen 3.8 Max 88.7%
- LVBench In use Hours-long video understanding with multiple-choice questions on reasoning, grounding and summarization.
Gemini 4 Argon 91.7%
- MMVU In use 3,000 expert-annotated questions on videos across 27 disciplines. Multiple-choice and open-ended accuracy.
Qwen 3.8 Max 82.4%
- MLVU In use Multi-task long video understanding (minutes to hours) with multiple-choice and generation tasks. Accuracy.
Qwen 3.8 Max 90.8%
- FLEURS In use Models transcribe read speech in 102 languages built from FLoRes-101 sentences. Model cards usually report word error rate, so lower scores are better.
Phi-4-multimodal 4.00%
- VoiceBench In use Spoken instructions with speaker, environment and content variation. Voice assistant responses scored by GPT judge or rules.
Qwen3.5-Omni-Plus 93.1%
Data audit
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 6 months ago, in May 2026, for Gemini 3.5 Flash. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (20 entries).
- Pass Models scored 4 models we track, plus 16 other entries on its leaderboard.
- Warning Newest score The newest came 6 months ago, in May 2026, for Gemini 3.5 Flash.
- Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3.5 Flash (#44).
- Note Headroom Lower is better, so there’s no ceiling to reach.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 5 rows for models we track and 16 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.