All Projects → morrolinux → Simple Ehm

morrolinux / Simple Ehm

Licence: mit
A simple tool for a simple task: remove filler sounds ("ehm") from pre-recorded speeches. AI powered.

Projects that are alternatives of or similar to Simple Ehm

Learnpythonforresearch
This repository provides everything you need to get started with Python for (social science) research.
Stars: ✭ 163 (-0.61%)
Mutual labels:  jupyter-notebook
Dl Course
Deep Learning with Catalyst
Stars: ✭ 162 (-1.22%)
Mutual labels:  jupyter-notebook
Tdl
Course "Theories of Deep Learning"
Stars: ✭ 164 (+0%)
Mutual labels:  jupyter-notebook
Neuralnets
Deep Learning libraries tested on images and time series
Stars: ✭ 163 (-0.61%)
Mutual labels:  jupyter-notebook
Machine Learning
Machine learning library written in readable python code
Stars: ✭ 163 (-0.61%)
Mutual labels:  jupyter-notebook
Workshops
Workshops organized to introduce students to security, AI, AR/VR, hardware and software
Stars: ✭ 162 (-1.22%)
Mutual labels:  jupyter-notebook
Ai Toolkit Iot Edge
AI Toolkit for Azure IoT Edge
Stars: ✭ 163 (-0.61%)
Mutual labels:  jupyter-notebook
Appearance Flow
A deep learning framework for synthesizing novel views of objects and scenes
Stars: ✭ 164 (+0%)
Mutual labels:  jupyter-notebook
Tensorflow2 Crash Course
A quick crash course in understanding the essentials of TensorFlow 2 and the integrated Keras API
Stars: ✭ 162 (-1.22%)
Mutual labels:  jupyter-notebook
Maml Jax
Implementation of Model-Agnostic Meta-Learning (MAML) in Jax
Stars: ✭ 164 (+0%)
Mutual labels:  jupyter-notebook
Notes
Contains Example Programs and Notebooks for some courses at Bogazici University, Department of Computer Engineering
Stars: ✭ 163 (-0.61%)
Mutual labels:  jupyter-notebook
Blog posts
Blog posts for matatat.org
Stars: ✭ 163 (-0.61%)
Mutual labels:  jupyter-notebook
Ta
Technical Analysis Library using Pandas and Numpy
Stars: ✭ 2,649 (+1515.24%)
Mutual labels:  jupyter-notebook
Imagecompletion Dcgan
Image completion using deep convolutional generative adversarial nets in tensorflow
Stars: ✭ 163 (-0.61%)
Mutual labels:  jupyter-notebook
Replika Research
Replika.ai Research Papers, Posters, Slides & Datasets
Stars: ✭ 164 (+0%)
Mutual labels:  jupyter-notebook
Data Science Template
A starter template for Equinor data science / data engineering projects
Stars: ✭ 163 (-0.61%)
Mutual labels:  jupyter-notebook
Csharp with csharpfritz
Show notes, slides, and samples from the CSharp with CSharpFritz show
Stars: ✭ 164 (+0%)
Mutual labels:  jupyter-notebook
Dsp
Metis Data Science Bootcamp - Official Prework Repository
Stars: ✭ 164 (+0%)
Mutual labels:  jupyter-notebook
Newsrecommender
A news recommendation system tailored for user communities
Stars: ✭ 164 (+0%)
Mutual labels:  jupyter-notebook
Competition Baseline
数据科学竞赛知识、代码、思路
Stars: ✭ 2,553 (+1456.71%)
Mutual labels:  jupyter-notebook

simple-ehm

A simple tool for a simple task: remove filler sounds ("ehm") from pre-recorded speeches. AI powered. Istruzioni in italiano in fondo al documento.

Usage

Basic invokation should be enough: ./simple_emh-runnable.py /path/to/video/file This will generate a subtilte track (.srt) for debugging and the output video in the same folder as the original file.

For more info read the help: ./simple_emh-runnable.py --help

Contributing to the model

There are two ways you can contribute to the model:

Contribute to the dataset

By sending me at least 30 1-second long WAV pcm_s16le mono 16kHz clips for each class (silence, speech, ehm) [easy]

  • You can convert your clips to the right format with ffmpeg: ffmpeg -i input-file -c:a pcm_s16le -ac 1 -ar 16000 -filter:a "dynaudnorm" output.wav
  • You can extract ehm(s) and silences along with erroneously classified sounds (false positives) by passing --generate-training-data as an invocation parameter. You can then use the latter to improve your training set!

Contribute to the training

  • By implementing transfer training logic on this model's python notebook
  • By retraining the current model with your dataset and make a PR with the updated one

ITA

simple-ehm

Un semplice strumento per un semplice compito: rimuovere gli "ehm" (suoni di riempimento) da discorsi pre-registrati.

Utilizzo

L'invocazione base dovrebbe essere sufficiente: ./simple_emh-runnable.py /percorso/al/file/video Questo genererò una traccia di sottotitoli (.srt) per fini diagnostici e il video tagliato nella stessa cartella del file originale.

Per maggiori informazioni sui parametri accettati, leggi la guida: ./simple_emh-runnable.py --help

Contribuire al modello

Ci sono due modi in cui puoi contribuire al modello:

Contribuisci al dataset

Inviandomi almeno 30 clip in formato WAV (pcm_s16le) mono con campionamento a 16kHz per ciascuna classe (silenzio, parlato, ehm) [facile]

  • Puoi convertire le tue clip nel formato corretto con ffmpeg: ffmpeg -i input-file -c:a pcm_s16le -ac 1 -ar 16000 -filter:a "dynaudnorm" output.wav
  • Puoi estrarre gli ehm(s) e i silenzi anche quelli classificati erroneamente (falsi positivi) passando --generate-training-data come parametro di invocazione. Puoi usare le clip classificate erroneamente per migliorare il tuo training set!

Contribuisci al training

  • Implementando la logica di transfer training sul notebook python di questo modello, e
  • Eseguendo il retraining della rete esistente con il tuo dataset ed inviandomi il modello aggiornato.
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].