faster-whisper
faster-whisper is an AI tool in the Voice & Audio category. Useful for speech-to-text, transcription, whisper, open-source, python.
OpenAI's Whisper reimplemented on CTranslate2: same accuracy, up to 4x faster, with word-level timestamps.
O Whisper da OpenAI reescrito em cima do CTranslate2: mesma precisão, até 4 vezes mais rápido, com o tempo de cada palavra.
AITag Review
faster-whisper is SYSTRAN's reimplementation of OpenAI's Whisper on CTranslate2, a fast inference engine for Transformer models: up to 4 times faster than the original for the same accuracy, using less memory, and lighter still with 8-bit quantization on CPU or GPU. One flag returns every word with its start and end time, and a built-in Silero VAD skips the silence. Models go from tiny (75 MB) to large-v3 (2.9 GB) and download themselves from Hugging Face on first use. We use it here to time every caption of our films, on the CPU, with the small model. MIT license.
Features
- Category: Voice & Audio
- Tags:speech-to-texttranscriptionwhisperopen-sourcepython
- Website:Visit
- Curated: No
Related tools
Voice synthesis platform that employs deep learning models to generate realistic human voices. Supports multiple languages, voice cloning based on provided samples, and detailed customization of voice parameters.
High-quality text-to-speech and voice cloning platform offering studio-grade audio with nuanced emotion control. Featuring over 2 million voices across 8 languages, it enables creators to generate professional narrations and realistic vocal replicas for any project.
Flow makes writing quick and clear with seamless voice dictation. It is the fastest, smartest way to type with your voice.
Music creation platform that uses AI to generate complete tracks, including vocals, instrumentation, and production. Offers controls to personalize the musical generation process.