Model reference · open weights

ai-light-dance_drums_ft_pretrain_wav2vec2

Available as managed deployment Audio gary109 · community Speech→text 1 variants 724 dl/mo

ai-light-dance_drums_ft_pretrain_wav2vec2 is an open-weight audio or speech model from gary109. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.

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

What it is

Released bygary109
TypeAudio & music
TaskSpeech→text
Runs withtransformers
Released2022-11-07
Popularity724 downloads / month
LicenceUnknown

About

What ai-light-dance_drums_ft_pretrain_wav2vec2 is

This model is a fine-tuned version of gary109/ai-light-dance_drums_ft_pretrain_wav2vec2-base on the GARY109/AI_LIGHT_DANCE - ONSET-DRUMS dataset. It achieves the following results on the evaluation set:

  • Loss: 1.8991
  • Wer: 0.6046
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: 0.0004
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 20
  • num_epochs: 200.0
  • mixed_precision_training: Native AMP

Training results

Training LossEpochStepValidation LossWer
No log0.982.04340.6226
0.47391.9162.10240.6247
0.46932.9241.98240.6211
0.51393.9322.29620.6429
0.50814.9402.22010.6292
0.50815.9482.13990.6208
0.57856.9562.14510.6417
0.5337.9642.11840.6330
0.51418.9722.02300.6342
0.49719.9802.21370.6381
0.497110.9882.11590.6253
0.564511.9962.09660.6247
0.493212.91041.92490.6223
0.491813.91122.04450.6235
0.505314.91202.13170.6304
0.505315.91282.07230.6256
0.556516.91362.13900.6402
0.481917.91441.95560.6321
0.513118.91521.98860.6333
0.479819.91601.97000.6259
0.479820.91681.97710.6295
0.522121.91761.98800.6235
0.486222.91842.09940.6298
0.483123.91922.05210.6205
0.495224.92001.98380.6064
0.495225.92082.03190.6103
0.511926.92162.04190.6160
0.499627.92242.00730.6178
0.48828.92322.17400.6304
0.497829.92402.27310.6163
0.497830.92482.24200.6205
0.525931.92562.05610.6184
0.4732.92641.94550.6136
0.513233.92721.93070.6043
0.497234.92802.05360.6127
0.497235.92881.91130.6223
0.514736.92961.93170.6286
0.491437.93042.18100.6241
0.47238.93122.14030.6160
0.482539.93202.11410.6094
0.482540.93282.28700.6031
0.513841.93362.14040.6181
0.4842.93442.02430.6265
0.459843.93522.11170.6199
0.47444.93602.03780.6321
0.47445.93682.19190.6211
0.493346.93762.36450.6109
0.469247.93842.19200.6076
0.471648.93922.36630.6034
0.460149.94002.28380.6280
0.460150.94082.02870.6148
0.489151.94162.13460.6130
0.450652.94242.15560.6181
0.458153.94322.05600.6229
0.448554.94401.99440.5971
0.448555.94481.97910.6097
0.494256.94562.11660.6070
0.474857.94642.02710.6124
0.422958.94722.04370.6229
0.4559.94802.10120.6142
0.4560.94881.91510.6049
0.493661.94961.89910.6046
0.460262.95041.98130.6112
0.462663.95121.93720.6136
0.44564.95201.90600.6154
0.44565.95281.95740.6151
0.490766.95362.09470.6022
0.472367.95442.00610.6010
0.410368.95521.95570.6094
0.480869.95602.10420.6088
0.480870.95682.13600.6073
0.468271.95762.12900.6013
0.447272.95841.94540.5989
0.425973.95922.09370.6043
0.446474.96002.08220.6058
0.446475.96082.01280.6058
0.477576.96161.97440.6094
0.439477.96241.99920.6010
0.41878.96322.16930.5947
0.438479.96402.13260.5923
0.438480.96482.11510.5950
0.497181.96562.15810.5923
0.417682.9664

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

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys ai-light-dance-drums-ft-pretrain-wav2vec2 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (ai-light-dance-drums-ft-pretrain-wav2vec2 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="ai-light-dance-drums-ft-pretrain-wav2vec2" -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.

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