Mongolian Speech-to-Text Leaderboard
ChimegeOperating point: stt-long (async submit+poll)

Chimege — Mongolian Speech-to-Text Benchmark

Real results from the Chimege 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 14, 2026, 4:15 PMEndpoint:https://api.chimege.com/v1.2/stt-long
06710085.4%
Accuracy

265/265 samples transcribed · 100% success rate

06710014.6%
Word Error Rate

Lower is better · across 265 samples

Headline metrics
Character Error Rate6.6%
Avg speed factor1.51×realtime multiple
Total speed factor1.51×
Avg latency / sample4.7s
Total audio processed1849.2s30.8 min

Pricing

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

Per 1k minutes
$11.1
batch
Per minute
$0.0111
effective
Per 1k min (these 30.8 min)
$20.53
would cost
Open source?
proprietary

Pricing source: Chimege public pricing. Duudlaga Flow is shown for context only — this page isolates Chimege 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 Chimege numbers, which are weak on Common Voice 24.

Common Voice 24 (MN)
Source dataset →
WER
10.7%
CER
4.4%
Shunya Labs Mongolian Speech
Source dataset →
WER
13.4%
CER
4.8%
Common Voice 20 (MN)
Source dataset →
WER
9.4%
CER
3.7%
Modern Voice
WER
21.1%
CER
10.8%
WERCER

Dataset summary

Aggregate accuracy, speed, and timing for each dataset.

DatasetSourceSamplesSuccessWERCERAccuracySpeedAudio (s)Proc (s)
Common Voice 24 (MN)Hugging Face →5959/5910.7%4.4%89.3%1.22×330.5s271.3s
Shunya Labs Mongolian SpeechHugging Face →6060/6013.4%4.8%86.6%2.17×645.1s296.8s
Common Voice 20 (MN)Hugging Face →5454/549.4%3.7%90.6%1.11×269.7s243.5s
Modern Voice9292/9221.1%10.8%78.9%1.46×603.8s413.5s

WER vs speed — per sample

Each dot is one audio sample. The sweet spot is the bottom-left (low error, fast). Chimege 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 Chimege 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)Chimege resultWERCERI/D/S
1btsee_0001Common Voice 24 (MN)Гэхдээ амьсгал хураахаасаа өмнө танд мэдэж байгаагаа хэлье.Гэхдээ амьсгал хураахаасаа өмнө танд мэдэж байгаагаа хэл.12.5%3.4%0/0/1
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 үү?0.0%0.0%0/0/0
3btsee_0003Common Voice 24 (MN)Одоо бид өөрсдөө өвчин эмгэгээсээ салахыг хичээцгээе.Одоо бид өөрсдөө өвчин эмгэгээсээ салахыг хичээцгээе.0.0%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 цээр дэр хийж явдаг.20.0%11.8%0/1/1
5btsee_0005Common Voice 24 (MN)Хан хурмаст уурлаж, Болдоггүй Бор өвгөнийг хор луугаараа ниргүүлэхээр явуулжээ.Хан хурмаст уурлаж болдоггүй бор өвгөн 2 луугаар эргүүлэхээр явжээ.50.0%16.9%0/0/5
6btsee_0006Common Voice 24 (MN)Алив наашаа ороод ир гээд гэртээ оров.Алив наашаа ороод ир гээд гэртээ ор.14.3%5.4%0/0/1
7btsee_0007Common Voice 24 (MN)Өө өндөр дээдэс таны тухайд би баталж чадахгүй.Өө өндөр дээдэс таны тухайд би баталж чадахгүй.0.0%0.0%0/0/0
8btsee_0008Common Voice 24 (MN)Харин гурав дахь удаагаас эхлэн хүмүүсийг сонирхож эхлэв.Харин 3 дахь удаагаасаа эхлэн хүмүүсийг сонирхож эхлэв.25.0%12.5%0/0/2
9btsee_0009Common Voice 24 (MN)Та нар очингуутаа шөл л өгч үз.Та нар очингуутаа шөл л өгч үз.0.0%0.0%0/0/0
10btsee_0010Common Voice 24 (MN)Ерөөсөө литр үйлдвэрээсээ салаагүй явсан юм чинь.Ерөөсөө л үйлдвэрээсээ салаагүй явсан юм чинь.14.3%6.2%0/0/1
Page 1 of 27

Methodology

How these numbers were produced.

Provider: Chimege (Chimege (Чимэгэ) Mongolian speech-to-text via the /v1.2/transcribe endpoint. Mongolian-native model, synchronous short-form transcription.).

Endpoint https://api.chimege.com/v1.2/stt-long. Language mn. Operating point stt-long (async submit+poll). Diarization none.

Datasets: Common Voice 24 (MN), Shunya Labs Mongolian Speech, Common Voice 20 (MN), Modern Voice — 265 samples, 1849.2s 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 Chimege 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 Chimege Benchmark Runner · Chimege Batch API v2 · run Aug 14, 2026, 4:15 PM