GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
aunikud
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
aunikud
Model Tree
aunikud/whisper-he-ipa
Input Modalities
Output Modalities
Supported Functionality
| Metric | Value |
|---|---|
| WER | 9.96% |
| CER | 3.31% |
data/ilspeech-v2/test, 150 samples)| Metric | Base (whisper-he-ipa) | This model |
|---|---|---|
| CER | 2.17% | 1.73% |
| WER | 9.55% | 7.99% |
| SER | 8.22% | 6.63% |
| VER | 1.97% | 1.69% |
| Exact match | 44.7% | 46.7% |
data/michael-gold-v1, 561 samples)| Metric | Base (whisper-he-ipa) | This model |
|---|---|---|
| CER | 4.20% | 4.05% |
| WER | 19.71% | 19.12% |
| SER | 14.55% | 14.98% |
| VER | 2.79% | 2.80% |
| Exact match | 6.6% | 7.7% |
Full per-sample reports are in benchmarks/.
python
from transformers import pipeline pipe = pipeline( "automatic-speech-recognition", model="aunikud/whisper-heb-ipa", generate_kwargs={"language": "he", "task": "transcribe"},)print(pipe("audio.wav")["text"])CLI from this repo:
bash
uv run src/infer.py audio.wav --model aunikud/whisper-heb-ipadata/ilspeech-v2/traindata/imaginary-jail-clean-v2 train split (metadata_train.csv, ~11h)Eval during training used metadata_eval_30m.csv (30-minute holdout from the same dataset).
from transformers import pipeline pipe = pipeline( "automatic-speech-recognition", model="aunikud/whisper-heb-ipa", generate_kwargs={"language": "he", "task": "transcribe"},)print(pipe("audio.wav")["text"])uv run src/infer.py audio.wav --model aunikud/whisper-heb-ipa