FaceMin

whisper-kss-model

Deploy Dedicated
Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more
Container

Run this model inference with full control and performance in your environment.

Learn more

Get help setting up a custom Dedicated Endpoints.

Talk with our engineer to get a quote for reserved GPU instances with discounts.

README

License: apache-2.0

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: 32
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 5000
  • mixed_precision_training: Native AMP

Training results

Table
Training LossEpochStepValidation LossWer
8.84490.0346257.2246100.6831
5.790.0692504.9938100.6516
4.14090.1037753.0934170.1858
2.71610.13831002.6763100.0528
2.70330.17291252.5222101.9024
2.58080.20751502.4298118.5847
2.37610.24201752.3257140.0440
2.31610.27662002.227391.3945
2.14910.31122252.224379.8649
2.09480.34582502.0493440.3429
1.71440.38042751.2596206.7145
1.11490.41493001.1042216.9204
1.07890.44953251.0707351.0901
1.02540.48413501.0166236.3858
1.080.51873750.9928191.0787
1.01160.55334000.9832166.7260
1.01050.58784250.9782165.3540
0.91220.62244500.9595117.0455
0.95770.65704750.9387118.7912
0.93310.69165000.9090138.5781
0.87570.72615250.9211137.1420
0.89260.76075500.8667184.8980
0.9120.79535750.8793212.2682
0.86910.82996000.8762166.9460
0.82630.86456250.8687154.8995
0.78060.89906500.8227156.4442
0.76450.93366750.8382216.6598
0.79750.96827000.7901165.6084
0.79341.00287250.7438120.6291
0.7691.03737500.7438218.1116
0.79921.07197750.7885259.8376
0.77451.10658000.7686179.9579
0.73491.14118250.7466162.1403
0.7251.17578500.7193218.4822
0.75561.21028750.7174209.3269
0.73951.24489000.7290291.2190
0.68181.27949250.6959212.7851
0.73251.31409500.6805186.9867
0.70341.34859750.6926215.1493
0.72541.383110000.6952195.1569
0.65741.417710250.6794190.6886
0.67281.452310500.6707201.2204
0.70471.486910750.6808212.6877
0.62621.521411000.6513265.1156
0.65881.556011250.6495451.8608
0.65211.590611500.6237353.7313
0.62451.625211750.6181292.6338
0.55291.659812000.5895221.0098
0.58951.694312250.5896176.8251
0.60881.728912500.5862124.4094
0.61041.763512750.6013149.8396
0.55981.798113000.5346107.1293
0.56951.832613250.5958145.3336
0.58291.867213500.5294141.4980
0.52031.901813750.5402235.1803
0.52061.936414000.5394230.0891
0.54481.971014250.5440214.9932
0.51792.005514500.5342255.9379
0.49532.040114750.5169231.7033
0.522.074715000.4929334.5504
0.49232.109315250.4894215.9817
0.4782.143815500.5068299.2691
0.49522.178415750.4833324.2779
0.48062.213016000.4889241.6753
0.49092.247616250.4777264.4795
0.48022.282216500.4487199.4435
0.43012.316716750.4488182.4179
0.46982.351317000.4612262.3237
0.47522.385917250.4646372.9149
0.43782.420517500.4663237.8527
0.44732.455017750.4687282.1577
0.46322.489618000.4266283.8381
0.40352.524218250.4152270.4255
0.44732.558818500.4201315.6571
0.41712.593418750.4169379.2287
0.39212.627919000.4186399.1616
0.44032.662519250.4181335.2756
0.40492.697119500.4180279.9728
0.4272.731719750.4203241.3959
0.44682.766320000.4185275.8122
0.38642.800820250.4206333.1377
0.42172.835420500.4207272.2071
0.45542.870020750.4220362.6389
0.38562.904621000.3459366.1595
0.31122.939121250.3450381.3861
0.39792.973721500.3453476.1855
0.34763.008321750.3465361.8760
0.30033.042922000.3488399.7811
0.3533.077522250.3473583.0224
0.33013.112022500.3489623.8666
0.31813.146622750.3495496.9255
0.35233.181223000.3513434.4554
0.35773.215823250.3515267.3580
0.33763.250323500.3539208.7754
0.36043.284923750.3508284.9401
0.31923.319524000.3529345.7634
0.33153.354124250.3508320.4169
0.33523.388724500.3510353.3194
0.26773.423224750.2803288.2148
0.26443.457825000.2801308.2338
0.2753.492425250.2818381.4079
0.27233.527025500.2835352.1826
0.27813.561525750.2836276.9052
0.26493.596126000.2854303.5409
0.29163.630726250.2862320.8857
0.30213.665326500.2873272.8570
0.3023.699926750.2864405.3588
0.27913.734427000.2866360.0782
0.26823.769027250.2870266.1347
0.28313.803627500.2894342.1414
0.28953.838227750.2901354.2226
0.31373.872828000.2916334.0133
0.2723.907328250.2903327.9886
0.28923.941928500.2898285.9423
0.31563.976528750.2898256.1146
0.24924.011129000.2870247.8596
0.18424.045629250.2905267.4770
0.19084.080229500.2905280.4566
0.20354.114829750.2927283.1202
0.20724.149430000.1993302.1461
0.16454.184030250.2001443.5101
0.21444.218530500.2008522.6579
0.21684.253130750.2025454.2819
0.20594.287731000.2025438.9496
0.2094.322331250.2036402.1667
0.20544.356831500.2030482.2121
0.25374.391431750.2037564.0528
0.19534.426032000.2024386.3496
0.24124.460632250.2047416.8283
0.21834.495232500.2046360.2043
0.23574.529732750.2044374.4222
0.1954.564333000.2043464.0528
0.20294.598933250.2042375.7738
0.1844.633533500.2043326.0008
0.19384.668033750.2041373.6381
0.1974.702634000.2032391.5910
0.19064.737234250.2050494.8205
0.22364.771834500.2047535.1733
0.20544.806434750.2048460.1114
0.1564.840935000.1438402.3349
0.14384.875535250.1477394.8726
0.12944.910135500.1389351.3879
0.14084.944735750.1354367.8489
0.1414.979336000.1345469.6171
0.14715.013836250.1347465.4007
0.13445.048436500.1346465.6937
0.14365.083036750.1345549.9895
0.14045.117637000.1346553.1492
0.14245.152137250.1348491.0546
0.15815.186737500.1353511.9900
0.14845.221337750.1351429.5145
0.16425.255938000.1358460.6926
0.13135.290538250.1358510.2741
0.12385.325038500.1360491.1906
0.14375.359638750.1363452.1762
0.13755.394239000.1371439.6003
0.15145.428839250.1365433.0299
0.12435.463339500.1370397.3949
0.14655.497939750.1364452.3959
0.14275.532540000.1363520.0983
0.13735.567140250.1365602.1553
0.15415.601740500.1368602.0611
0.10275.636240750.1205571.3285
0.09245.670841000.1206597.3635
0.11675.705441250.1210631.2152
0.11395.740041500.1207663.5221
0.10385.774641750.1206612.3378
0.10565.809142000.1209660.6076
0.10295.843742250.1207622.7394
0.11575.878342500.1208633.6045
0.10955.912942750.1212599.1143
0.10245.947443000.1211589.3203
0.1015.982043250.1212560.9578
0.08076.016643500.1213562.8218
0.09586.051243750.1213579.9176
0.10026.085844000.1212634.5932
0.11446.120344250.1212635.9835
0.09546.154944500.1210633.5839
0.08316.189544750.1214578.2492
0.08356.224145000.1214604.1504
0.10796.258645250.1213629.2688
0.09636.293245500.1214615.5510
0.09026.327845750.1214603.1823
0.08846.362446000.0835600.7927
0.07756.397046250.0838601.6594
0.08566.431546500.0840585.1707
0.09416.466146750.0840586.3968
0.08936.500747000.0839610.2315
0.08616.535347250.0838593.9224
0.08286.569847500.0838590.7726
0.09576.604447750.0838540.1015
0.0916.639048000.0837529.6480
0.09346.673648250.0837552.8063
0.08786.708248500.0837585.4244
0.08416.742748750.0837608.0013
0.10346.777349000.0837603.1603
0.09186.811949250.0837621.2134
0.08366.846549500.0837620.0507
0.08266.881149750.0836615.6115
0.08286.915650000.0836612.9690

Framework versions

  • Transformers 4.57.1
  • Pytorch 2.8.0+cu126
  • Datasets 4.4.2
  • Tokenizers 0.22.1

Model provider

FaceMin

Model tree

Base

openai/whisper-small

Fine-tuned

this model

Modalities

Input

Audio

Output

Text

Pricing

Dedicated Endpoints

View details

Supported Functionality

Model APIs

Dedicated Endpoints

Container

More information

Model card

Explore FriendliAI today

Get startedTalk to an engineer