---
language:
- kk
- ru
- en
tags:
- transformers
- text-generation
- automatic-speech-recognition
- multimodal
- gemma-3n
- asr
- kazakh
- trl
- lora
- unsloth
pipeline_tag: audio-text-to-text
library_name: transformers
model-index:
- name: Estimin3n
  results:
  - task:
      type: automatic-speech-recognition
      name: ASR (audio→text)
    metrics:
    - type: wer
      value: N/A
      name: Word Error Rate
    - type: cer
      value: N/A
      name: Character Error Rate
  - task:
      type: text-generation
      name: Text generation (chat)
    metrics: []
base_model:
- google/gemma-3n-E4B-it
license: gemma
---

![Screenshot 2025-10-02 at 01-11-34 pitch deck НОВЫЙ ДИЗАЙН 24.09 - pitch deck НОВЫЙ ДИЗАЙН 24.09 (2).pdf](https://cdn-uploads.huggingface.co/production/uploads/65435de3c9483f8861b424ce/lsFz1xhQHrlnrJkcYbQ20.png)

# Estimin3n - Opensource multimodal kazakh audio/text to text LLM 
[**github**](https://github.com/govnejri/Estimin3n)

- audio (16 kHz, mono) and/or text as input,
- generating text responses or transcriptions.


Backbone: Gemma 3N. Input formatting is done via `AutoProcessor.apply_chat_template`. This repository contains full weights (`safetensors`) and processor/tokenizer configs.

Hardware support is provided in partnership with [**ait🍅maton**](https://huggingface.co/aitomaton)

## Terms of Use: [Terms](https://ai.google.dev/gemma/terms)
## Benchmark Results

# Audio

Word error rate (WER) over the KSC2-test and other benchmarks :


![image](https://cdn-uploads.huggingface.co/production/uploads/65435de3c9483f8861b424ce/S1sfaoPo6NTQY2WzBuWVL.png)


# Text


![image](https://cdn-uploads.huggingface.co/production/uploads/65435de3c9483f8861b424ce/QmnJTvTpCT1q1nizcpVod.png)


## Installation

```bash
pip install -U transformers accelerate soundfile librosa safetensors
# Optional for 4‑bit loading
pip install -U bitsandbytes
```

Recent `transformers` is recommended (>= 4.44).

## Quickstart (Python)

Replace `your-username/Estimin3n` with your model repo id.


### Audio transcription (Kazakh)

```python
import soundfile as sf
from transformers import AutoProcessor, Gemma3nForConditionalGeneration
import torch

repo_id = "your-username/Estimin3n"
device = "cuda" if torch.cuda.is_available() else "cpu"
dtype = torch.bfloat16 if device == "cuda" else torch.float32

model = Gemma3nForConditionalGeneration.from_pretrained(
    repo_id, trust_remote_code=True, torch_dtype=dtype, device_map="auto"
)
processor = AutoProcessor.from_pretrained(repo_id, trust_remote_code=True)

audio, sr = sf.read("sample.wav")
if audio.ndim > 1:
    audio = audio.mean(axis=1)
audio = audio.astype("float32")

messages = [
    {"role": "system", "content": [{"type": "text", "text": (
        "You are an expert assistant that accurately transcribes Kazakh speech. "
        "Output clean Kazakh text only."
    )}]},
    {"role": "user", "content": [
        {"type": "audio", "audio": audio},
        {"type": "text", "text": "Transcribe this audio."}
    ]},
]

inputs = processor.apply_chat_template(
    messages, add_generation_prompt=True, tokenize=True,
    return_dict=True, return_tensors="pt"
).to(device)

if "input_features" in inputs:
    inputs["input_features"] = inputs["input_features"].to(dtype)
for k in ["input_ids", "attention_mask"]:
    if k in inputs:
        inputs[k] = inputs[k].long()

with torch.no_grad():
    out = model.generate(
        **inputs, max_new_tokens=256, do_sample=False,
        pad_token_id=processor.tokenizer.eos_token_id,
        eos_token_id=processor.tokenizer.eos_token_id,
    )
gen = out[0][inputs["input_ids"].shape[1]:]
print(processor.tokenizer.decode(gen, skip_special_tokens=True).strip())
```

### 3) Text → text (no audio)

```python
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

repo_id = "your-username/Estimin3n"
tok = AutoTokenizer.from_pretrained(repo_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    repo_id, trust_remote_code=True, torch_dtype=torch.bfloat16, device_map="auto"
)

prompt = "Қазақ тілінде қысқаша сәлемдесу мәтінін жазыңыз."
inputs = tok(prompt, return_tensors="pt").to(model.device)
with torch.no_grad():
    out = model.generate(**inputs, max_new_tokens=64, do_sample=True, temperature=0.7)
print(tok.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))
```

## Loading options

- bfloat16/float16 on ≥24GB VRAM is recommended for best quality.
- 4‑bit (`bitsandbytes`) reduces VRAM to ~8–12GB at some quality cost.

```python
from transformers import Gemma3nForConditionalGeneration
model = Gemma3nForConditionalGeneration.from_pretrained(
  "your-username/Estimin3n",
  trust_remote_code=True,
  device_map="auto",
  load_in_4bit=True,
)
```

## Training summary

- Base: `unsloth/gemma-3n-E4B-it`
- Method: SFT (TRL `SFTTrainer`) with LoRA
- LoRA: `r=8`, `lora_alpha=16`, `lora_dropout=0`, targets in attention/MLP and audio submodules
- Example SFT params: `per_device_train_batch_size=1`, `gradient_accumulation_steps=2`, `lr=5e-5`, `epochs>=1`, `cosine`, `weight_decay=0.01`, `remove_unused_columns=False`, `dataset_text_field="messages"`, `max_seq_length=2048`
- Data: custom Kazakh audio corpus (FLAC + transcripts) converted to HF `DatasetDict`

## Evaluation

Companion scripts provide evaluation pipelines:
- WER/CER on your dataset (logs to TSV with detailed samples)
- KazMMLU multiple‑choice benchmark (Kazakh/Russian subsets)

Public benchmark scores will be added in upcoming releases.

## Limitations & Safety

- The model can generate inaccurate or biased outputs; human review is advised.
- For audio inputs, ensure user privacy and consent.
- Quality depends on recording conditions (noise, accent, speed).


## Acknowledgements

- Google Gemma 3N; Hugging Face `transformers` / `datasets` / `trl`
- Unsloth (efficient fine‑tuning & quantization)
- KazMMLU community
- ait🍅maton - https://huggingface.co/aitomaton

### Citations

If this project or its results are useful, please cite the KSC2 dataset paper:

```
@inproceedings{mussakhojayeva22_interspeech,
  title     = {KSC2: An Industrial-Scale Open-Source Kazakh Speech Corpus},
  author    = {Saida Mussakhojayeva and Yerbolat Khassanov and Huseyin {Atakan Varol}},
  year      = {2022},
  booktitle = {Interspeech 2022},
  pages     = {1367--1371},
  doi       = {10.21437/Interspeech.2022-421},
  issn      = {2958-1796},
}
```

```
@article{gemma_3n_2025,
    title={Gemma 3n},
    url={https://ai.google.dev/gemma/docs/gemma-3n},
    publisher={Google DeepMind},
    author={Gemma Team},
    year={2025}
}
```
tsu