Model reference · open weights

uzbek-stt-3

Available as managed deployment Audio sarahai · community Speech→text 1 variants 4k dl/mo

uzbek-stt-3 is an open-weight audio or speech model from sarahai. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.

Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.

What it is

Released bysarahai
TypeAudio & music
TaskSpeech→text
Parameters (lead)315M
Runs withtransformers
Released2024-08-13
Popularity4k downloads / month
LicenceOpen weights

About

What uzbek-stt-3 is

This model is a fine-tuned version of oyqiz/uzbek_stt based mainly on laws and military related dataset.

Read the full model card

Model Details

Model Description

  • Developed by: Sara Musaeva
  • Funded by: SSD
  • Model type: Transformers
  • Language(s) (NLP): Uzbek
  • Finetuned from model: Oyqiz/uzbek-stt

Model Sources

Uses

Intended for Speech-to-text conversion

How to Get Started with the Model

from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC
import torch
import torchaudio

model_name = "sarahai/uzbek-stt-3"
model = Wav2Vec2ForCTC.from_pretrained(model_name)
processor = Wav2Vec2Processor.from_pretrained(model_name)

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model.to(device)

def load_and_preprocess_audio(file_path):
    speech_array, sampling_rate = torchaudio.load(file_path)
    if sampling_rate != 16000:
        resampler = torchaudio.transforms.Resample(orig_freq=sampling_rate, new_freq=16000)
        speech_array = resampler(speech_array)
    return speech_array.squeeze().numpy()

def replace_unk(transcription):
    return transcription.replace("[UNK]", "ʼ")

audio_file = "/content/audio_2024-08-13_15-20-53.ogg"
speech_array = load_and_preprocess_audio(audio_file)

input_values = processor(speech_array, sampling_rate=16000, return_tensors="pt").input_values.to(device)

with torch.no_grad():
    logits = model(input_values).logits

predicted_ids = torch.argmax(logits, dim=-1)
transcription = processor.batch_decode(predicted_ids)

transcription_text = replace_unk(transcription[0])

print("Transcription:", transcription_text)

From the published model card. Full card on the HuggingFace links in the sidebar.

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys uzbek-stt-3 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (uzbek-stt-3 below is illustrative; you get the exact model name on deployment.)

$ curl -sS https://api.axforge.ai/v1/audio/transcriptions \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -F model="uzbek-stt-3" -F file=@audio.mp3

Create an account — your API key is available in the console. 3M free tokens every 30 days with every new account.

© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms