TTS-Grabber

Quick thing i made about a year ago to download any text with any tts voice, over 630 voices to choose from currently.

It will split the input into multiple files every 1500 words or so to not hit any cutoff limits from TTS providers.

Usage:

Edit input.txt to change the text to synthesize.

You can run just tts.py without any parameters to open the voice selector with default settings.

Parameters

PARAMETER           TYPE  DESCRIPTION
-h, -help           ---   Shows the help info.
-v, -voice          Int   Sets the voice id to use.  
-s, -speed          Int   Sets the TTS voice speed (in percent).  
-vol, -volume       Int   Changes the TTS volume (in decibels).  
-pp, -period-pause  Flt   Sets how long the TTS should pause for at periods (in seconds).  
-cp, -comma-pause   Flt   Sets how long the TTS should pause for at commas (in seconds).  
-lp, -line-pause    Flt   Sets how long the TTS should pause for at newlines (in seconds).

Example with parameters:
tts.py -v 184 -s 100 -vol 0 -pp 1 -cp 0.5 -lp 2

Every Google, Azure & IBM text to speech voice for free

Related tags

Overview

TTS-Grabber

Usage:

Parameters

absolutely no api abuse here

Owner

A Word Level Transformer layer based on PyTorch and 🤗 Transformers.

Textpipe: clean and extract metadata from text

Source code of the "Graph-Bert: Only Attention is Needed for Learning Graph Representations" paper

GraphNLI: A Graph-based Natural Language Inference Model for Polarity Prediction in Online Debates

BeautyNet is an AI powered model which can tell you whether you're beautiful or not.

Learning General Purpose Distributed Sentence Representations via Large Scale Multi-task Learning

Pytorch-version BERT-flow: One can apply BERT-flow to any PLM within Pytorch framework.

PortaSpeech - PyTorch Implementation

Library for Russian imprecise rhymes generation

NLP-based analysis of poor Chinese movie reviews on Douban

A repository to run gpt-j-6b on low vram machines (4.2 gb minimum vram for 2000 token context, 3.5 gb for 1000 token context). Model loading takes 12gb free ram.

Code for ACL 2020 paper "Rigid Formats Controlled Text Generation"

JaQuAD: Japanese Question Answering Dataset

Create a semantic search engine with a neural network (i.e. BERT) whose knowledge base can be updated

ChatBotProyect - This is an unfinished project about a simple chatbot.

Semi-automated vocabulary generation from semantic vector models

Quantifiers and Negations in RE Documents

Signature remover is a NLP based solution which removes email signatures from the rest of the text.

code for modular summarization work published in ACL2021 by Krishna et al