[ Date Index ]
[ Thread Index ]
[ <= Previous by date /
thread ]
[ Next by date /
thread => ]
[LUG]voice to text
- To: Devon/Cornwall GNU LUG <list@xxxxxxxxxxxxx>
- Subject: [LUG]voice to text
- From: Tom via list <list@xxxxxxxxxxxxx>
- Date: Fri, 7 Apr 2023 09:33:23 +0100
- Cc: Tom <madtom1999@xxxxxxxxxxxxxx>
- Content-language: en-GB
- Delivered-to: dclug@xxxxxxxxxxxxxxxxxxxxx
- Dkim-signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=dcglug.org.uk; s=1665479163; h=Content-Type:Cc:From:List-Unsubscribe: List-Subscribe:List-Post:List-Owner:List-Help:List-Id:Subject:Reply-To:To: MIME-Version:Date:Message-ID:Sender:Content-Transfer-Encoding:Content-ID: Content-Description:Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc :Resent-Message-ID:In-Reply-To:References:List-Archive; bh=8t3YvCeOAw6K3jSY0g7UX+WcAuhROkrjYp5EdQ4+eWA=; b=U2AhdQCC9J0tF7j6DMhpyAlo1x HwCJXMlZ4KBzQjwHZxJqcpBacqmpa5H41nelt1Cucss9jHj5fOPulF+Ha7dyBSdfNLhNB9kDUn/HA ULLtY3KPx2ufeF/iQ9n6QxEuHkmqxiIvrNmZRy2PEvqALompLymtSUxW/7ASxGy4/4hk=;
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=20210112; t=1680856546; x=1683448546; h=subject:from:to:content-language:user-agent:mime-version:date :message-id:from:to:cc:subject:date:message-id:reply-to; bh=85+aVGzHRh1ixC176wId32/73kzVEgtd9ozOE6h2S8w=; b=Tn0ydYsXw76qn2MYJsiKfodf7o9kPRVE/R8epIVaev3E5+/VHlM3wRkO03ZycjIHWF QbgGf1ZEJWILEkopL2gevvnL1T9u0rwm7wSIFVElBJUnrM+psyys1SKCvsNp8MA65eJX 8rXhOgvbLXW7E6IyS08jZHJ5/oYUmhEi4otFqu4htHHLwIZS05hdbVWNZ9h3X/hBnELx Pm1W5SzYST6rA/hXaR3Ik4UjHMFtJW4Ah595T+ISug8Af5zTeAUZ522gbbqWADALSl7A 2RyFmlvgptujFh/AoCGgbLJOYarzY3adFB1L8BKtwCCxT+q1mv6C1TaFFB4Cb2b3nLGZ emAA==
Just reading the latest Pimag (128) and it mentioned an offline
voice to text app and I've had a play with the small model (uses
<500meg ram) and it works quite well - Grandma is deaf and her
hearing aids spend more time in the repair shop than her ears so
I've been looking for something to help and my voice tends to
confuse these things - though there it would appear there is a way
of making your own models but not tried that yet.
https://github.com/ggerganov/whisper.cpp
built by
git clone https://github.com/ggerganov/whisper.cpp
cd ,/whisper.cpp
sudo apt install libsdl2-dev
./models/download-ggml-model.sh tiny.en
make -j tiny.en
make -j stream
run
./stream -m models/ggml-tiny.en.bin --step 4000 --length 8000 -c 0
-t 4 -ac 512
seemed to get the tv in the background near 100%
Tom te tom te tom
--
The Mailing List for the Devon & Cornwall LUG
FAQ: https://www.dcglug.org.uk/faq/
