ytrbqrkflbvbhy/whisper-podlodka-turbo-ct2-float16

CTranslate2 Conversion of bond005/whisper-podlodka-turbo

This repository contains a CTranslate2-converted version of the bond005/whisper-podlodka-turbo model, optimized for use with the faster-whisper library.

This is a full conversion, including all necessary configuration files (tokenizer.json, preprocessor_config.json, etc.) for maximum compatibility.

Model Details

• Original Model: bond005/whisper-podlodka-turbo (fine-tuned version of Whisper-Large-V3-Turbo)
• Format: CTranslate2
• Quantization: float16

This conversion was performed to enable high-performance inference on both CPU and GPU using faster-whisper.

Disclaimer

This is only a conversion of the original model. All credits for training and architecture go to the original author, bond005.