GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Creaturelove7
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
Creaturelove7
Model Tree
openai/whisper-small
Input Modalities
Output Modalities
Supported Functionality
Evaluated on 11,919 held-out test utterances (the official test split with all dev files removed — the pinned dataset revision ships every dev file inside test, and the evaluation excludes them):
| Model | CER | Mixed-language ER |
|---|---|---|
| whisper-small (base) | 25.53% | 18.61% |
| whisper-small + this adapter | 8.51% | 7.88% |
That is a 67% relative CER reduction. On an in-domain dev subset, the adapted 244M model also outperformed the 6x larger whisper-large-v3 served by a cloud API (CER 4.29% vs 9.46%). This is an in-domain result: on out-of-domain audio the general-purpose large model is expected to win.
The output is normalized Simplified Chinese with English technical terms preserved in Latin script, and the normalization survives int8 quantization.
Full evaluation reports (aggregate metrics, pinned revisions, package versions, hashed sample identifiers — no transcripts) are in the training repository.
python
from peft import PeftModelfrom transformers import WhisperForConditionalGeneration, WhisperProcessorbase = "openai/whisper-small"processor = WhisperProcessor.from_pretrained(base, language="zh", task="transcribe")model = WhisperForConditionalGeneration.from_pretrained(base)model = PeftModel.from_pretrained(model, "Creaturelove7/whisper-small-lora-ntu-ml2021")model.generation_config.forced_decoder_ids = processor.get_decoder_prompt_ids(language="zh", task="transcribe")
The decoder prompt is fixed to Chinese transcription, matching training; English technical terms are still emitted in Latin script. Use model.merge_and_unload() for a standalone checkpoint.
openai/whisper-small @ 973afd24965f72e36ca33b3055d56a652f456b4dky552/ML2021_ASR_ST @ 1e121cc419e87eed7d4825400baa06f102931944, all 17,779 official train utterances, 1 epochq_proj/v_proj (<1% trainable parameters)The complete, revision-pinned training and evaluation pipeline (hash-locked dependencies, split-leakage guards, Colab notebook) lives in training/ntu_ml2021.
The training corpus is the public Hugging Face dataset ky552/ML2021_ASR_ST (NTU ML2021 course lecture audio; the dataset card declares MIT). The underlying lecture recordings are the work of the course authors; this repository redistributes no audio, transcripts, or per-sample predictions — only LoRA weight deltas. If you are a rights holder of the source material and want this adapter removed, open an issue and it will be taken down.