Zum Inhalt springen
ModelAufgelistetExtern
Herunterladen

Whisper Tiny Es Mix Norm

This model is a fine-tuned version of openai/whisper-tiny on the following datasets: deepdml/commonvoice260 disco-eth/WorldSpeech facebook/voxpopuli facebook/multilinguallibrispeech deepdml/voxforge google/fleurs...

Quellehuggingface.co/deepdml/whisper-tiny-es-mix-norm/tree/c967c037e8bfa9f7b4db9119dae5faf0c0ed86eb
ArtText modelBasismodellopenai/whisper-tinyVersionmodelDBewertung

Modellquelle

Art
Text model
Dateigröße
144 MB
Basismodell
openai/whisper-tiny
Version
vmodel
Aktualisiert
28. Okt. 2026
Downloads
194
Quelle
Hugging Face

Quellenbeschreibung

Whisper Tiny es

This model is a fine-tuned version of openai/whisper-tiny on the following datasets:

  • deepdml/common_voice_26_0
  • disco-eth/WorldSpeech
  • facebook/voxpopuli
  • facebook/multilingual_librispeech
  • deepdml/voxforge
  • google/fleurs
  • deepdml/basque_parliament_1

It achieves the following results on the evaluation set:

  • Best checkpoint: 44,000
  • Loss: 0.2888
  • Wer: 16.0693
  • Cer: 6.0413

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: 128
  • eval_batch_size: 128
  • seed: 42
  • optimizer: Use adamw_torch 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: 0.04
  • training_steps: 46000

Training results

Training LossEpochStepValidation LossWer RawCer RawWerCer
0.36630.021710000.597131.872212.147831.651012.1036
0.26250.043520000.514028.278910.556728.234510.5484
0.21830.065230000.476325.38939.274525.36839.2709
0.23200.087040000.443524.30339.007324.29449.0057
0.32900.108750000.436323.96739.111723.95909.1103
0.20300.130460000.411822.94288.574122.93908.5735
0.34560.152270000.413122.71908.658322.71398.6574
0.32840.173980000.399622.67528.456722.67278.4563
0.23360.195790000.369220.92367.928820.92237.9286
0.29500.2174100000.362221.14368.278821.14368.2788
0.38750.2391110000.357520.11977.546020.11977.5460
0.18841.0020120000.340619.39067.230819.39067.2308
0.13701.0237130000.337319.32157.362519.32157.3625
0.14261.0455140000.336419.12377.190319.12377.1903
0.16161.0672150000.332918.57856.959218.57856.9592
0.15601.0889160000.327819.09527.397719.09527.3977
0.15111.1107170000.330818.78777.280318.78777.2803
0.16421.1324180000.327618.35736.782318.35736.7823
0.24951.1542190000.330018.61087.057918.61087.0579
0.19031.1759200000.323218.45047.056118.45047.0561
0.19821.1976210000.316118.36367.071018.36367.0710
0.38911.2194220000.315418.52407.178718.52407.1787
0.27251.2411230000.315517.82606.692017.82606.6920
0.12712.0040240000.301617.15406.564617.15406.5646
0.11712.0257250000.309017.27006.555017.27006.5550
0.17382.0474260000.303717.37216.601517.37216.6015
0.15322.0692270000.306917.19776.433017.19776.4330
0.22202.0909280000.305617.41396.597917.41396.5979
0.15092.1127290000.301016.99236.529716.99236.5297
0.14842.1344300000.303316.78636.208416.78636.2084
0.15582.1561310000.302117.14076.501617.14076.5016
0.15632.1779320000.297916.66776.223616.66776.2236
0.28232.1996330000.300617.18506.491217.18506.4912
0.14632.2213340000.295516.60186.306816.60186.3068
0.12092.2431350000.291616.63736.395016.63736.3950
0.11453.0059360000.294116.88016.504816.88016.5048
0.15453.0277370000.292516.62086.409516.62086.4095
0.32113.0494380000.290516.63546.496616.63546.4966
0.19143.0712390000.292616.26456.148716.26456.1487
0.16723.0929400000.293616.55176.282816.55176.2828
0.12563.1146410000.291016.68616.522016.68616.5220
0.10233.1364420000.291816.41796.237616.41796.2376
0.15973.1581430000.291316.61516.425716.61516.4257
0.19123.1798440000.288816.06936.041316.06936.0413
0.14593.2016450000.289116.23796.110716.23796.1107
0.24853.2233460000.288916.34506.185216.34506.1852

Framework versions

  • Transformers 5.14.1
  • Pytorch 2.6.0+cu124
  • Datasets 5.0.1
  • Tokenizers 0.22.2

Citation

Please cite the model using the following BibTeX entry:

@misc{deepdml/whisper-tiny-es-mix-norm,
      title={Fine-tuned Whisper tiny ASR model for speech recognition in Spanish},
      author={Jimenez, David},
      howpublished={\url{https://huggingface.co/deepdml/whisper-tiny-es-mix-norm}},
      year={2026}
    }

Quellen

1 Quelle
Hugging FacePrimärVersion c967c037e8bfa9f7b4db9119dae5faf0c0ed86eb
huggingface.co/deepdml/whisper-tiny-es-mix-norm/tree/c967c037e8bfa9f7b4db9119dae5faf0c0ed86eb

Verifiziert 28. Okt.

Modellartefakte

1 Artefakt

model.safetensors

safetensors · 144 MB · SHA-256 95d6c5b5026c…9399 · Hugging Face

Download file

Quellenauszüge

3 Auszüge
Readme28. Okt.huggingface.co/deepdml/whisper-tiny-es-mix-norm

--- library_name: transformers language: - es license: apache-2.0 base_model: openai/whisper-tiny tags: - asr datasets: - deepdml/common_voice_26_0 - disco-eth/WorldSpeech - facebook/voxpopuli - facebook/multilingual_librispeech - deepdml/voxforge - google/fleurs - deepdml/basque_parliament_1 metrics: - wer model-index: - name: Whisper Tiny es results: - task: name: Automatic Speech Recognition type: automatic-speech-recognition dataset: name: Common Voice 26.0 type: deepdml/common_voice_26_0 metrics: - name: Wer type: wer value: 16.345032205710808 pipeline_tag: automatic-speech-recognition --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # Whisper Tiny es This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the following datasets: - deepdml/common_voice_26_0 - disco-eth/WorldSpeech - facebook/voxpopuli - facebook/multilingual_librispeech - deepdml/voxforge - google/fleurs - deepdml/basque_parliament_1 It achieves the following results on the evaluation set: - Best checkpoint: 44,000 - Loss: 0.2888 - Wer: 16.0693 - Cer: 6.0413 ## 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: 128 - eval_batch_size: 128 - seed: 42 - optimizer: Use adamw_torch 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: 0.04 - training_steps: 46000 ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer Raw | Cer Raw | Wer | Cer | |:-------------:|:------:|:-----:|:---------------:|:-------:|:-------:|:-------:|:-------:| | 0.3663 | 0.0217 | 1000 | 0.5971 | 31.8722 | 12.1478 | 31.6510 | 12.1036 | | 0.2625 | 0.0435 | 2000 | 0.5140 | 28.2789 | 10.5567 | 28.2345 | 10.5484 | | 0.2183 | 0.0652 | 3000 | 0.4763 | 25.3893 | 9.2745 | 25.3683 | 9.2709 | | 0.2320 | 0.0870 | 4000 | 0.4435 | 24.3033 | 9.0073 | 24.2944 | 9.0057 | | 0.3290 | 0.1087 | 5000 | 0.4363 | 23.9673 | 9.1117 | 23.9590 | 9.1103 | | 0.2030...

Readme9. Aug.huggingface.co/deepdml/whisper-tiny-es-mix-norm

--- library_name: transformers language: - es license: apache-2.0 base_model: openai/whisper-tiny tags: - generated_from_trainer datasets: - deepdml/voxforge - fixie-ai/common_voice_17_0 - google/fleurs - deepdml/basque_parliament_1 - facebook/multilingual_librispeech - facebook/voxpopuli metrics: - wer model-index: - name: Whisper Tiny es results: - task: name: Automatic Speech Recognition type: automatic-speech-recognition dataset: name: Common Voice 17.0 type: deepdml/voxforge metrics: - name: Wer type: wer value: 21.674748489027575 --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # Whisper Tiny es This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set: - Loss: 0.3749 - Wer Raw: 21.6741 - Cer Raw: 8.4585 - Wer: 21.6747 - Cer: 8.4585 ## 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: 128 - eval_batch_size: 128 - seed: 42 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - lr_scheduler_warmup_ratio: 0.04 - training_steps: 20000 ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer Raw | Cer Raw | Wer | Cer | |:-------------:|:------:|:-----:|:---------------:|:-------:|:-------:|:-------:|:-------:| | 0.3891 | 0.05 | 1000 | 0.6939 | 37.6291 | 14.8103 | 37.4810 | 14.7830 | | 0.6597 | 0.1 | 2000 | 0.6883 | 34.8302 | 14.4460 | 34.8016 | 14.4405 | | 0.1388 | 0.15 | 3000 | 0.6530 | 34.3358 | 13.1348 | 34.2976 | 13.1269 | | 0.1217 | 0.2 | 4000 | 0.6682 | 34.5982 | 13.2295 | 34.5646 | 13.2233 | | 0.1101 | 0.25 | 5000 | 0.6721 | 35.2325 | 13.7696 | 35.1969 | 13.7628 | | 0.3564 | 0.3 | 6000 | 0.5374 | 28.6885 | 11.2075 | 28.6529 | 11.2010 | | 0.3803 | 0.35 | 7000 | 0.3890 | 20.9394 | 7.6348 | 20.9369 | 7.6342 | | 0.3373 | 0.4 | 8000 | 0.3540 | 19.9264 | 7.3914 | 19.9258 | 7.3913 | | 0.2828 |...

Note28. Okt.huggingface.co/deepdml/whisper-tiny-es-mix-norm

Source context: 194 downloads · 2 likes · Pipeline automatic-speech-recognition · Library transformers · Repo deepdml/whisper-tiny-es-mix-norm