GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
maleo-ai
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
maleo-ai
Model Tree
openai/whisper-base
Input Modalities
Output Modalities
Supported Functionality
WER / CER (%), full test sets, lowercase + punctuation stripped:
| Test set | whisper-base | whisper-base-id | whisper-base-id, short context |
|---|---|---|---|
| FLEURS id (684 utts) | 38.8 / 12.5 | 15.0 / 4.7 | 15.2 / 4.8 |
| Common Voice 17 id (3629 utts) | 43.1 / 19.1 | 14.5 / 5.0 | 14.5 / 5.0 |
Latency, whisper.cpp q8_0, 4 s utterance:
| CPU | Settings | ms / utterance | RTF |
|---|---|---|---|
| Cortex-A53 x4 1.4 GHz (Orange Pi Zero3) | -t 4 -ac 256 | 1870 | 0.47 |
| RK3588, Cortex-A76 x4 (Orange Pi 5 Plus) | taskset -c 4-7, -t 4 -ac 256 | 198 | 0.05 |
Output is lowercase without punctuation.
model.safetensors (fp16) + tokenizer/config: transformersggml-base-id-q8_0.bin (82 MB): whisper.cpp. On ARM, q8_0 beats q5_0 on speed.bash
# -ac = encoder positions: 50 per second of audio, round up to a multiple of 64 (4 s -> 256, 8 s -> 448)whisper-cli -m ggml-base-id-q8_0.bin -l id -t 4 -ac 256 -bs 1 -bo 1 -nt -f audio16k.wavpython
from transformers import pipelineasr = pipeline("automatic-speech-recognition", model="maleo-ai/whisper-base-id")print(asr("audio.wav", generate_kwargs={"language": "indonesian", "task": "transcribe"})["text"])bibtex
@misc{irawan2026whisperbaseid, author = {Eka Tresna Irawan}, title = {whisper-base-id: Whisper base fine-tuned for Indonesian speech recognition}, year = {2026}, publisher = {Hugging Face}, howpublished = {\url{https://huggingface.co/maleo-ai/whisper-base-id}}}# -ac = encoder positions: 50 per second of audio, round up to a multiple of 64 (4 s -> 256, 8 s -> 448)whisper-cli -m ggml-base-id-q8_0.bin -l id -t 4 -ac 256 -bs 1 -bo 1 -nt -f audio16k.wavfrom transformers import pipelineasr = pipeline("automatic-speech-recognition", model="maleo-ai/whisper-base-id")print(asr("audio.wav", generate_kwargs={"language": "indonesian", "task": "transcribe"})["text"])@misc{irawan2026whisperbaseid, author = {Eka Tresna Irawan}, title = {whisper-base-id: Whisper base fine-tuned for Indonesian speech recognition}, year = {2026}, publisher = {Hugging Face}, howpublished = {\url{https://huggingface.co/maleo-ai/whisper-base-id}}}