Phowhisper Medium
PhoWhisper is a series of models designed specifically for Vietnamese automatic speech recognition (ASR). It achieves high robustness by fine-tuning the Whisper model on an 844-hour Vietnamese accent dataset.
Downloads 2,999
Release Time : 2/18/2024
Model Overview
PhoWhisper offers five versions, focusing on the Vietnamese automatic speech recognition task and achieving the current state-of-the-art performance on Vietnamese ASR benchmark datasets.
Model Features
Multi-accent adaptation
Trained on a dataset containing 844 hours of different Vietnamese accents, with strong accent robustness
State-of-the-art performance
Achieves the current state-of-the-art performance on Vietnamese ASR benchmark datasets
Multiple version options
Provides five different versions of the model to meet different needs
Model Capabilities
Vietnamese speech recognition
Multi-accent speech processing
Use Cases
Speech transcription
Vietnamese meeting records
Automatically transcribe Vietnamese meeting recordings into text
Highly accurate transcription results
Media subtitle generation
Automatically generate subtitles for Vietnamese video content
Support subtitle generation for multiple Vietnamese accents
Featured Recommended AI Models
Š 2025AIbase