Тёмный

Whisper Medusa - Speech Recognition Model - Beats OpenAI Whisper - Install Locally 

Fahd Mirza
Подписаться 16 тыс.
Просмотров 1,6 тыс.
50% 1

This video shows how to install whisper-medusa-v1 which builds on Whisper by predicting multiple tokens per iteration, which significantly improves speed.
🔥 Buy Me a Coffee to support the channel: ko-fi.com/fahd...
🔥 Get 50% Discount on any A6000 or A5000 GPU rental, use following link and coupon:
bit.ly/fahd-mirza
Coupon code: FahdMirza
▶ Become a Patron 🔥 - / fahdmirza
#medusa #whisper
PLEASE FOLLOW ME:
▶ LinkedIn: / fahdmirza
▶ RU-vid: / @fahdmirza
▶ Blog: www.fahdmirza.com
RELATED VIDEOS:
▶ Resource huggingface.co...
All rights reserved © 2021 Fahd Mirza

Опубликовано:

 

8 сен 2024

Поделиться:

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист
Посмотреть позже
Комментарии : 7   
@Kaalkian
@Kaalkian Месяц назад
how does it compare to WhisperX?
@SaddamBinSyed
@SaddamBinSyed Месяц назад
Hi @Fahd, thank you for this video. I wanted to ask if you have tried using microphone input with the model? Also, how about streaming STT support? As you know, the Whisper model can process up to 30s audio chunks. Is this limitation present here as well? Thanks
@deepharia4209
@deepharia4209 Месяц назад
How did got hands of that gpu compute unit in server which website?
@Gerald-iz7mv
@Gerald-iz7mv Месяц назад
Can you do real time?
@reinerzufall3123
@reinerzufall3123 Месяц назад
listen at 1:00
@Gerald-iz7mv
@Gerald-iz7mv Месяц назад
@@reinerzufall3123 faster than almost realtime…
@reinerzufall3123
@reinerzufall3123 Месяц назад
@@Gerald-iz7mv if you need it faster i guess you have to stick to closed source models 😊
Далее
This Open Source Scraper CHANGES the Game!!!
20:36
Просмотров 52 тыс.
Нарвался на сотрудника ФСБ⚡️
01:00
Can Whisper be used for real-time streaming ASR?
8:41
host ALL your AI locally
24:20
Просмотров 1 млн
Run your own AI (but private)
22:13
Просмотров 1,4 млн
We Need to Rethink Exercise - The Workout Paradox
12:00
INSANELY FAST Talking AI: Powered by Groq & Deepgram
12:11