GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
maqsudxo1ja
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
maqsudxo1ja
Model Tree
maqsudxo1ja/uz-whisper-small-stt
Input Modalities
Output Modalities
Supported Functionality
| Ko'rsatkich | Qiymat |
|---|---|
| WER (Word Error Rate) | 28.64% |
| Validation Loss | 0.2634 |
| Model hajmi | ~242M parametr (Whisper-small) |
| Training Loss | Epoch | Step | Validation Loss | WER |
|---|---|---|---|---|
| 3.0847 | 0.68 | 400 | 0.3161 | 32.49 |
| 2.4038 | 1.35 | 800 | 0.2827 | 29.58 |
| 2.0104 | 2.03 | 1200 | 0.2708 | 30.53 |
| 1.9056 | 2.70 | 1600 | 0.2647 | 28.92 |
| 1.9695 | 3.00 | 1776 | 0.2634 | 28.64 |
bash
pip install transformers torch librosa soundfilepipeline bilan (eng oson)python
from transformers import pipeline transcriber = pipeline( "automatic-speech-recognition", model="maqsudxo1ja/uz-whisper-small-stt-v2") result = transcriber("audio_fayl.mp3")print(result["text"])python
import torchimport librosafrom transformers import WhisperProcessor, WhisperForConditionalGeneration MODEL_ID = "maqsudxo1ja/uz-whisper-small-stt-v2" processor = WhisperProcessor.from_pretrained(MODEL_ID, language="uzbek", task="transcribe")model = WhisperForConditionalGeneration.from_pretrained(MODEL_ID) device = "cuda" if torch.cuda.is_available() else "cpu"model = model.to(device) # Audio faylni 16kHz formatida yuklaymiz (Whisper shuni talab qiladi)audio, sr = librosa.load("audio_fayl.mp3", sr=16000) input_features = processor( audio, sampling_rate=sr, return_tensors="pt").input_features.to(device) predicted_ids = model.generate(input_features, language="uzbek", task="transcribe")transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)[0] print(transcription)python
import sounddevice as sdimport numpy as npfrom transformers import pipeline transcriber = pipeline("automatic-speech-recognition", model="maqsudxo1ja/uz-whisper-small-stt-v2") duration = 5 # soniyaprint("🎙️ Gapiring...")audio = sd.rec(int(duration * 16000), samplerate=16000, channels=1, dtype="float32")sd.wait() result = transcriber({"array": audio.flatten(), "sampling_rate": 16000})print("Natija:", result["text"])Asosiy model: maqsudxo1ja/uz-whisper-small-stt
Dataset: islomov/it_youtube_uzbek_speech_dataset — 21,000 audio klip, IT mavzusidagi YouTube videolardan (Mohir Dev)
| Parametr | Qiymat |
|---|---|
| Learning rate | 1e-5 |
| Train batch size | 4 (effektiv: 32, gradient accumulation orqali) |
| Optimizer | AdamW (fused) |
| LR scheduler | linear, 200 warmup step |
| Epochs | 3 |
Apache 2.0
maqsudxo1ja/uz-whisper-small-sttislomov/it_youtube_uzbek_speech_datasetfrom transformers import pipeline transcriber = pipeline( "automatic-speech-recognition", model="maqsudxo1ja/uz-whisper-small-stt-v2") result = transcriber("audio_fayl.mp3")print(result["text"])import torchimport librosafrom transformers import WhisperProcessor, WhisperForConditionalGeneration MODEL_ID = "maqsudxo1ja/uz-whisper-small-stt-v2" processor = WhisperProcessor.from_pretrained(MODEL_ID, language="uzbek", task="transcribe")model = WhisperForConditionalGeneration.from_pretrained(MODEL_ID) device = "cuda" if torch.cuda.is_available() else "cpu"model = model.to(device) # Audio faylni 16kHz formatida yuklaymiz (Whisper shuni talab qiladi)audio, sr = librosa.load("audio_fayl.mp3", sr=16000) input_features = processor( audio, sampling_rate=sr, return_tensors="pt").input_features.to(device) predicted_ids = model.generate(input_features, language="uzbek", task="transcribe")transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)[0] print(transcription)import sounddevice as sdimport numpy as npfrom transformers import pipeline transcriber = pipeline("automatic-speech-recognition", model="maqsudxo1ja/uz-whisper-small-stt-v2") duration = 5 # soniyaprint("🎙️ Gapiring...")audio = sd.rec(int(duration * 16000), samplerate=16000, channels=1, dtype="float32")sd.wait() result = transcriber({"array": audio.flatten(), "sampling_rate": 16000})print("Natija:", result["text"])