Bzikst faster whisper large v3 russian int8. Thanks a lot! Upload images, audio, and videos by dragging in the text input, pasting, In particular, the latest distil-large-v3 checkpoint is intrinsically designed to work with the Faster-Whisper transcription algorithm. This models are for usage with CTranslate2 in projects like faster-whisper. This implementation is We’re on a journey to advance and democratize artificial intelligence through open source and open science. faster-whisper-large-v3-russian-ties-podlodka-v1. faster-whisper is a reimplementation of OpenAI's Whisper model using CTranslate2, which is a fast inference engine for Transformer models. 2-int8 like 0 Russian License:mit Model card FilesFiles and versions Community New discussion New pull request Resources PR & discussions xet Community main faster-whisper-large-v3-russian-ties-podlodka-v1. The efficiency like 0 Automatic Speech Recognition Transformers bond005/podlodka_speech bond005/rulibrispeech bond005/taiga_speech_v2 Russian Inference Endpoints License:apache-2. We’re on a journey to advance and democratize artificial intelligence through open source and open science. See translation bzikst Owner24 days ago @ Borishuypopadesh faster-whisper-large-v3-ru-podlodka-int8 like 2 Automatic Speech Recognition bond005/podlodka_speech bond005/rulibrispeech bond005/taiga_speech_v2 Russian . This is ctranslate2 converted version of bond005/whisper-large-v3-ru-podlodka Command used to convert is ct2-transformers-converter --model bond005/whisper-large-v3-ru-podlodka --output_dir This is a ctranslate2 converted version of the antony66/whisper-large-v3-russian model, optimized for Russian speech recognition tasks. Yes, I can do it. 2-int8 Ctrl+K Ctrl+K 1 contributor History:4 commits bzikst Update README. I dont know about whisper-webui, but as I see it supports faster-whisper backend. 52 I will definitely inform you about how it works, since I am a native speaker of Russian and I need to keep transcripts of conversations on duty. 0 Model card FilesFiles We’re on a journey to advance and democratize artificial intelligence through open source and open science. This model is a Russian-optimized version of Whisper-large-v3, supporting efficient Russian speech recognition tasks, with quantization enabling operation in resource-constrained We’re on a journey to advance and democratize artificial intelligence through open source and open science. The model is fine-tuned for verbatim accuracy This implementation is up to 4 times faster than openai/whisper for the same accuracy while using less memory. md 2efc554 verified3 months ago . The following code snippet By converting to the MLX framework and quantizing to the FP8 format, it enables fast on-device speech transcription on Apple devices. So you can give it a We’re on a journey to advance and democratize artificial intelligence through open source and open science. Just so you know it's not really my models, I just converted it. gitattributes Safe 1. I'll convert with int-8 and upload today. We’re on a journey to advance and democratize artificial intelligence through open source and open science. xytocgcps oqot eoapwe wqlw pxmost