protrolium

ffmpeg

Converting Audio into Different Formats / Sample Rates

Minimal example: transcode from MP3 to WMA:
ffmpeg -i input.mp3 output.wma

You can get the list of supported formats with:
ffmpeg -formats

You can get the list of installed codecs with:

harperreed

EMFIT Local API

There is a functional http server running on the device that can be crawled for realtime data. It runs on port 80 and uses plain http.

You will need to know the IP address if your emfit device.

shortdvm.htm

For demoing / testing with a web browser there is an auto-refreshing page that contains only HR and RR values, it’s name is shortdvm.htm

seidler2547

# prerequisites:
## install software
apt install mariadb-server libmariadbclient-dev sqlite3
## install mysqlclient in virtualenv
su -c 'homeassistant/bin/pip3 install mysqlclient --upgrade' -l homeassistant
## create database
mysql -e 'CREATE SCHEMA IF NOT EXISTS `hass_db` DEFAULT CHARACTER SET utf8'
## create user (use a safe password please)
mysql -e "CREATE USER 'hass_user'@'localhost' IDENTIFIED BY 'hass_pw'"
mysql -e "GRANT ALL PRIVILEGES ON hass_db.* TO 'hass_user'@'localhost'"

SteveALee

Build your own Audacity for Windows with ASIO driver support

You might want Audacity to work with the ASIO drivers supplied with your sound devices. Commmon reasons for this are:

• Only ASIO drivers are available for your hardware (eg Behringer mixers)
• Performance - ASIO has low latency (delays)
• You don't want to use the ASIO4All bridging driver with non ASIO drivers
• Multi channel support - though Audacity is not so good at handling other than Stereo or Mono

Due to licensing restrictions the Audacity team cannot provide a prebuilt version with Windows ASIO driver support. But with this guide and script you can easily build your own Audacity with ASIO support, on a Windows PC.

Shulyaka

#!/usr/bin/env python3
"""
This is an example of a team work of two OpenAI Assistants. They have been given a task to review a test of a new employee candidate. The 'Reviewer' (aka Junior hiring manager) is given the instructions to review the test, and the 'Supervisor' (aka Senior hiring manager) is given the instructions to guide the Reviewer throug the process, to ask him to review section by section, and to check his work. User and Reviewer only talk to Supervisor and not to each other.

As of Jan 2024, it is not perfect and the AI Agents do make mistakes and still require human intervention to complete the task correctly. Howewer this approach significantly reduces the number of interventions required and provides better results because of self-check. The downside is the number of tokens required is several times greater than with a single agent approach.

The instructions are very custom and thus not included. But here is an example:

```supervisor_instructions.md
You are a senior hiring manager of a te

lainedfles

#!/bin/bash

# ssh-cipher-benchmark.sh - Assesses speed of SSH encryption between specific hosts.
# Usage:
# ssh-cipher-benchmark.sh [script options] [cipher selection]
# Default ciphers: all we can find...
#
# Note: In some cases, the first cipher tested runs faster than the others, regardless of order.
# Cause of this is not known, but changing the order of testing shows it to be true.  Run the
# first one twice if you suspect this.  Perhaps it is due to buffering?