Instructions to use razhan/whisper-base-me with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use razhan/whisper-base-me with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="razhan/whisper-base-me")# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("razhan/whisper-base-me") model = AutoModelForSpeechSeq2Seq.from_pretrained("razhan/whisper-base-me", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download all_results.json from razhan/whisper-base-me: direct link, hf CLI and curl.
- Browser
- Download file 1.12 kB
-
https://huggingface.co/razhan/whisper-base-me/resolve/main/all_results.json
- Command line
-
hf download hf://razhan/whisper-base-me/all_results.json
-
curl -L -o all_results.json https://huggingface.co/razhan/whisper-base-me/resolve/main/all_results.json
1.12 kB
| { | |
| "epoch": 4.0, | |
| "eval_avg_cer": 0.2678389158869433, | |
| "eval_avg_wer": 0.7003291969579674, | |
| "eval_gilaki_cer": 0.38182211642501623, | |
| "eval_gilaki_wer": 0.9606072252113759, | |
| "eval_hawrami_cer": 0.08648390486197345, | |
| "eval_hawrami_wer": 0.4144927536231884, | |
| "eval_laki_kurdish_cer": 0.16528387831462502, | |
| "eval_laki_kurdish_wer": 0.603680981595092, | |
| "eval_loss": 0.5420735478401184, | |
| "eval_mazanderani_cer": 0.2510331798323297, | |
| "eval_mazanderani_wer": 0.6836474783494652, | |
| "eval_runtime": 458.9768, | |
| "eval_samples": 3258, | |
| "eval_samples_per_second": 7.098, | |
| "eval_southern_kurdish_cer": 0.17209076399972242, | |
| "eval_southern_kurdish_wer": 0.5609543712991989, | |
| "eval_steps_per_second": 0.057, | |
| "eval_talysh_cer": 0.5, | |
| "eval_talysh_wer": 0.9166666666666666, | |
| "eval_zazaki_cer": 0.31815856777493606, | |
| "eval_zazaki_wer": 0.7622549019607843, | |
| "total_flos": 5.41529079939072e+18, | |
| "train_loss": 0.6912094768954486, | |
| "train_runtime": 12274.8821, | |
| "train_samples": 20873, | |
| "train_samples_per_second": 6.802, | |
| "train_steps_per_second": 0.053 | |
| } |