AxForge
Sign in Get started

Model reference · open weights

whisper-large-mn-13

Available as managed deployment Audio bayartsogt · community Speech→text 1 variants 835 dl/mo

whisper-large-mn-13 is an open-weight audio or speech model from bayartsogt. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.

Request deployment All audio & music

Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.

What it is

Released bybayartsogt
TypeAudio & music
TaskSpeech→text
Runs withtransformers
Released2022-12-20
Popularity835 downloads / month
LicenceOpen weights

About

What whisper-large-mn-13 is

This model is a fine-tuned version of openai/whisper-large-v2 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1689
  • Wer: 20.0240
  • Cer: 6.6010
Read the full model card

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 1e-05
  • train_batch_size: 8
  • eval_batch_size: 4
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 25000
  • mixed_precision_training: Native AMP

Training results

Training LossEpochStepCerValidation LossWer
0.39210.09100015.78450.410146.9030
0.31150.17200014.29110.335341.8451
0.26590.26300011.81310.280034.6406
0.24770.35400010.66590.257832.0024
0.22740.43500010.04600.246330.3419
0.20590.5260009.92640.230528.5558
0.20920.6170009.42770.219627.8785
0.19560.6980009.27450.209326.8353
0.1950.7890008.94850.204226.6168
0.1950.87100008.53240.200125.6718
0.17950.95110008.17860.193624.1698
0.15751.04120007.86530.191523.8912
0.13581.13130007.67490.191823.3778
0.15091.21140007.72210.185223.1811
0.14741.3150007.32460.176422.4984
0.14611.39160007.31870.179322.4110
0.1341.47170007.11230.173721.9412
0.12891.56180007.45930.172722.0614
0.12871.65190007.02300.170121.4223
0.11961.73200006.94470.166621.2475
0.12751.82210006.79560.165320.8106
0.13291.91220006.77290.162220.3354
0.12941.99230006.64480.160620.2207
0.10432.08240006.60100.168920.0240
0.0792.17250006.62460.168720.1005

Framework versions

  • Transformers 4.26.0.dev0
  • Pytorch 1.13.1+cu117
  • Datasets 2.8.1.dev0
  • Tokenizers 0.13.2

From the published model card. Full card on the HuggingFace links in the sidebar.

Benchmarks

Reported results

As published on the model card — the maker's own numbers, not measured by AxForge.

TaskDatasetMetricScore
Automatic Speech RecognitionCommon Voice 11.0Wer20.024
Automatic Speech RecognitionCommon Voice 11.0Cer6.601

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys whisper-large-mn-13 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (whisper-large-mn-13 below is illustrative; you get the exact model name on deployment.)

$ curl -sS https://api.axforge.ai/v1/audio/transcriptions \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -F model="whisper-large-mn-13" -F file=@audio.mp3

Create an account — your API key is available in the console. 3M free tokens every 30 days with every new account.

Get API access
© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms