Mongolian Speech-to-Text Leaderboard
Moonshine MNOperating point: orgilj/moonshine-mn (Moonshine encoder-decoder) on mps

Moonshine MN — Mongolian Speech-to-Text Benchmark

Real results from the Moonshine MN Batch API (enhanced operating point, language mn) evaluated across 4 Mongolian datasets and 265 audio samples. Every number below is measured — not marketing. WER, speed, and pricing are shown as-is, brutally honest.

Language:MNDiarization:noneSamples:265Run:Aug 18, 2026, 5:30 PMEndpoint:local (transformers)
06710041.2%
Accuracy

265/265 samples transcribed · 100% success rate

06710058.8%
Word Error Rate

Lower is better · across 265 samples

Headline metrics
Character Error Rate45.2%
Avg speed factor8.44×realtime multiple
Total speed factor8.44×
Avg latency / sample0.8s
Total audio processed1849.3s30.8 min

Pricing

Moonshine MN list pricing for batch transcription. No discounts, no negotiated rates applied — the raw per-minute rate.

Per 1k minutes
$0
batch
Per minute
$0.0000
effective
Per 1k min (these 30.8 min)
$0.00
would cost
Open source?
proprietary

Pricing source: Moonshine MN public pricing. Duudlaga Flow is shown for context only — this page isolates Moonshine MN so the number is not padded by our own product.

WER & CER by dataset

Word and character error rates per dataset. Lower is better — and these are the real Moonshine MN numbers, which are weak on Common Voice 24.

Common Voice 24 (MN)
Source dataset →
WER
17.3%
CER
11.8%
Shunya Labs Mongolian Speech
Source dataset →
WER
94.3%
CER
72.6%
Common Voice 20 (MN)
Source dataset →
WER
2.7%
CER
1.8%
Modern Voice
WER
95.4%
CER
74.1%
WERCER

Dataset summary

Aggregate accuracy, speed, and timing for each dataset.

DatasetSourceSamplesSuccessWERCERAccuracySpeedAudio (s)Proc (s)
Common Voice 24 (MN)Hugging Face →5959/5917.3%11.8%82.7%7.37×330.6s44.9s
Shunya Labs Mongolian SpeechHugging Face →6060/6094.3%72.6%5.7%10.04×645.1s64.2s
Common Voice 20 (MN)Hugging Face →5454/542.7%1.8%97.3%7.33×269.8s36.8s
Modern Voice9292/9295.4%74.1%4.6%8.25×603.8s73.2s

WER vs speed — per sample

Each dot is one audio sample. The sweet spot is the bottom-left (low error, fast). Moonshine MN sits high on error for many Common Voice 24 samples.

Per-sample results

Ground truth shown verbatim in the Expected column. The result column highlights only the words Moonshine MN got wrong, in red — no strikethrough/swap gymnastics, just the mistakes.

265 rows
redwrong word in the resultplaincorrectly transcribedExpected column shown verbatim as ground truth
#AudioSampleDatasetExpected (ground truth)Moonshine MN resultWERCERI/D/S
1btsee_0001Common Voice 24 (MN)Гэхдээ амьсгал хураахаасаа өмнө танд мэдэж байгаагаа хэлье.гэхдээ амьсгал хураахаасаа өмнө танд мэдэж байгаагаа хэлье0.0%0.0%0/0/0
2btsee_0002Common Voice 24 (MN)Надад заяасан аз жаргал гэдэг ердөө гуравхан сарын хугацаатай байсан гэж үү?надад заяасан аз жаргал гэдэг ердөө гуравхан сарын хугацаатай байсан гэж үү0.0%0.0%0/0/0
3btsee_0003Common Voice 24 (MN)Одоо бид өөрсдөө өвчин эмгэгээсээ салахыг хичээцгээе.одоо бид өөрсдөө өвчин эмгэгээсээ салахыг хичээцгээе0.0%0.0%0/0/0
4btsee_0004Common Voice 24 (MN)Би бол голдуу хээрээр гэр, хэцээр дэр хийж явдаг хүн.би бол голдуу хээрээр гэр хэцээр дэр хийж явдаг хүн0.0%0.0%0/0/0
5btsee_0005Common Voice 24 (MN)Хан хурмаст уурлаж, Болдоггүй Бор өвгөнийг хор луугаараа ниргүүлэхээр явуулжээ.хан хурмаст уурлаж болдоггүй бор өвгөнийг хоёр луугаараа ниргүүлэхээр явуулжээ10.0%1.3%0/0/1
6btsee_0006Common Voice 24 (MN)Алив наашаа ороод ир гээд гэртээ оров.алив наашаа ороод ир гээд гэртээ оров0.0%0.0%0/0/0
7btsee_0007Common Voice 24 (MN)Өө өндөр дээдэс таны тухайд би баталж чадахгүй.өө өндөр дээдэс таны тухайд би баталж чадахгүй0.0%0.0%0/0/0
8btsee_0008Common Voice 24 (MN)Харин гурав дахь удаагаас эхлэн хүмүүсийг сонирхож эхлэв.харин гурав дахь удаагаас эхлэн хүмүүс сонирхож эхэлнэ25.0%10.7%0/0/2
9btsee_0009Common Voice 24 (MN)Та нар очингуутаа шөл л өгч үз.та нар очингуутаа шөл л өгч үз0.0%0.0%0/0/0
10btsee_0010Common Voice 24 (MN)Ерөөсөө литр үйлдвэрээсээ салаагүй явсан юм чинь.ерөөсөө л үйлдвэрээсээ салаагүй явсан юм чинь14.3%6.2%0/0/1
Page 1 of 27

Methodology

How these numbers were produced.

Provider: Moonshine MN (Moonshine — an edge-oriented ASR architecture with variable-length audio encoding — trained for Mongolian. 190 MB, the smallest model on the board.). Models with this badge were trained on data that overlaps the benchmark corpora (Mongolian Common Voice — 113 of 173 samples — or the public Shunya Labs corpus). Scores on overlapping corpora are inflated by memorization; judge these models by the per-dataset table on their details page, especially the corpora they were NOT trained on.

Endpoint local (transformers). Language mn. Operating point orgilj/moonshine-mn (Moonshine encoder-decoder) on mps. Diarization none.

Datasets: Common Voice 24 (MN), Shunya Labs Mongolian Speech, Common Voice 20 (MN), Modern Voice — 265 samples, 1849.3s of audio total.

Dataset source URLs:

Metrics: WER and CER are computed with a standard word/character Levenshtein alignment, normalized for case and punctuation. Accuracy = 100 − WER. Speed factor = audio duration ÷ processing time (× realtime). All requests are real Moonshine MN Batch API calls, not cached or simulated.

Diff highlighting: The result column aligns to the ground truth and colors every substitution and insertion red. Deletions (words missing from the result) are not shown in the result column — the Expected column already holds the full ground truth as-is.

Generated by the Moonshine MN Benchmark Runner · Moonshine MN Batch API v2 · run Aug 18, 2026, 5:30 PM