GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
DBGad
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
DBGad
Model Tree
Input Modalities
Output Modalities
Supported Functionality
This model is a fine-tuned version of OpenAI's Whisper model for Automatic Speech Recognition (ASR). It has been trained on a custom speech dataset to improve transcription accuracy for domain-specific speech and conversational audio.
The model leverages the strong multilingual capabilities of Whisper while adapting to the characteristics of the target dataset through supervised fine-tuning.
This model is designed for automatic speech recognition tasks, including:
The model can be integrated into:
Like all speech recognition models, performance depends on:
Potential limitations include:
python
from transformers import pipeline pipe = pipeline( "automatic-speech-recognition", model="DBGad/Meezan-ASR") result = pipe("sample.wav") print(result["text"])The following preprocessing steps were applied:
| Parameter | Value |
|---|---|
| Rank (r) | 8 |
| Alpha | 16 |
| Dropout | 0.1 |
Target Modules:
| Hyperparameter | Value |
|---|---|
| Base Model | openai/whisper-medium |
| Epochs | 12 |
| Batch Size (per device) | 16 |
| Gradient Accumulation | 8 |
| Effective Batch Size | 128 |
| Learning Rate | 5e-4 |
| Warmup Steps | 150 |
| Evaluation Strategy | Every 200 Steps |
| Checkpoint Saving | Every 200 Steps |
| Logging Steps | Every 50 Steps |
| Best Model Selection Metric | Word Error Rate (WER) |
| Load Best Model | Enabled |
The model was trained using:
These techniques significantly reduced GPU memory consumption while maintaining strong model performance.
Validation was performed every 200 training steps.
The best checkpoint was automatically selected using the lowest Word Error Rate (WER).
Evaluation Metric:
Training was performed on an NVIDIA GPU with CUDA support.
The objective of the model is to convert spoken English audio into accurate text transcriptions.
The model minimizes sequence-to-sequence cross-entropy loss while optimizing transcription quality measured by Word Error Rate (WER).
For questions, suggestions, or collaboration:
text
undefinedfrom transformers import pipeline pipe = pipeline( "automatic-speech-recognition", model="DBGad/Meezan-ASR") result = pipe("sample.wav") print(result["text"])