Faster Whisper Small.en is a converted version of the OpenAI Whisper small.en model adapted for the CTranslate2 inference engine. It supports English automatic speech recognition tasks and is hosted on Hugging Face by Systran under the MIT license.
The repository contains the model weights transformed from the original openai/whisper-small.en using the ct2-transformers-converter tool with float16 quantization. Model weights are stored in FP16 format, which can be adjusted at load time through the compute_type option. It is explicitly intended for use in CTranslate2 or any project built on CTranslate2, including the faster-whisper library.
Usage follows the faster-whisper Python API. Code examples load the model with WhisperModel("small.en"), call transcribe on an audio file such as audio.mp3, and iterate over resulting segments that provide start time, end time, and transcribed text. The conversion process also copies the tokenizer.json file into the output directory.
The model addresses efficient local inference for English speech-to-text on compatible hardware. It is delivered as a downloadable repository of model files on the Hugging Face platform rather than a standalone application.
In the Voice, TTS & speech space, Faster Whisper Small.en takes a focused approach. It focuses on running fast and efficient English speech-to-text transcription locally using optimized Whisper models. It is built as an open-source project for developers. Faster Whisper Small.en is open source under the MIT license. Faster Whisper Small.en is available on the web and the command line.
It is developed by Systran, and the product first shipped in 2018. Development happens publicly on GitHub with 4.6k stars and 21 commits in the last 90 days. It operates in a well-populated space: PulseGate tracks 7 similar tools. Key capabilities include Speech Recognition, CTranslate2 Format, and English Transcription.
Latest indexed changes and source events
Systran/faster-whisper-small.en verified by the PulseGate indexer
Other apps tracked under the same category.
