sebington’s gists

sebington / groq_language_detector.py

Last active October 30, 2025 10:51

Sends a character chain to a Groq model and retrieves the corresponding RFC 5646 language code (Groq API key needed)

	import os
	import json
	from groq import Groq

	def get_language_code(text: str, api_key: str = None) -> str:
	"""
	Identify the RFC 5646 language code of given text using Groq API.

	Args:
	text: The text to identify

sebington / new_machina_nn_v0.py

Created October 28, 2025 15:50

Code from the following YT video : https://www.youtube.com/watch?v=rdFjMb6dTJQ

	# =========================================================================
	# A simple PyTorch example that learns y = x^2
	# =========================================================================
	#!/usr/bin/env -S uv run
	# /// script
	# requires-python = ">=3.9"
	# dependencies = [
	# "torch",
	# "matplotlib",
	# "numpy",

sebington / r1-cpu.py

Last active February 8, 2025 16:25

CPU-friendly version of vgel's r1.py script (https://gist.github.com/vgel/8a2497dc45b1ded33287fa7bb6cc1adc)

	# Generated with Claude 3.5 Sonnet using vgel's r1.py script + the following prompt:
	# "Can you modify this script to improve inference speed on a CPU-only PC?"

	# It is possible to define the number of threads (= CPU cores) in the prompt
	# Example run : python r1-cpu.py -t 32 "What is 1+1?" --threads 4

	import argparse
	import random
	import sys
	from transformers import AutoModelForCausalLM, AutoTokenizer, DynamicCache

sebington / faster-whisper_word-level.ipynb

Last active February 1, 2025 01:36

Transcribe audio file at word-level and write output to .srt

Sorry, something went wrong. Reload?

Sorry, we cannot display this file.

Sorry, this file is invalid so it cannot be displayed.

sebington / batch_whisper.ipynb

Created December 27, 2024 14:58

Batch transcribe audio/video files with OpenAI Whisper

Sorry, something went wrong. Reload?

Sorry, we cannot display this file.

Sorry, this file is invalid so it cannot be displayed.

sebington / whisper_groq2srt_shorter.py

Last active March 25, 2025 19:52

Transcribe an audio/video file with Groq Whisper

	from groq import Groq
	from datetime import timedelta

	def format_time(seconds):
	# Ensure the first segment starts at 00:00:00,000
	if float(seconds) < 0.001:
	return "00:00:00,000"

	time = str(timedelta(seconds=float(seconds))).replace('.', ',')[:-3]
	return f"{time:0>12}"

sebington / whisper_groq2srt_claude.py

Last active February 1, 2025 01:36

Transcribe an audio/video file with Groq Whisper

	import os
	import math
	import tkinter as tk
	from tkinter import filedialog, messagebox
	from typing import Dict, Optional
	from groq import Groq
	import logging

	# Configure logging
	logging.basicConfig(

sebington / whisper_groq2srt.py

Last active March 30, 2025 12:34

Transcribe an audio/video file with Groq Whisper

	# /// script
	# requires-python = ">=3.13"
	# dependencies = [
	# "groq",
	# ]
	# ///
	from groq import Groq
	import math

	client = Groq()

sebington / whisper_batch_groq.ipynb

Created December 26, 2024 22:28

Batch transcribe audio/video files using Groq Whisper

Sorry, something went wrong. Reload?

Sorry, we cannot display this file.

Sorry, this file is invalid so it cannot be displayed.

sebington / batch_faster_whisper.ipynb

Created December 26, 2024 22:20

Batch transcribe audio/video files using Faster-Whisper

Sorry, something went wrong. Reload?

Sorry, we cannot display this file.

Sorry, this file is invalid so it cannot be displayed.

sebington sebington