Extend Vocabulary

emailmeemilyong · December 8, 2022, 7:45am

Please provide the following information when requesting support.

Hardware - GPU (A100/A30/T4/V100) A100
Hardware - CPU
Operating System: Ubuntu 20
Riva Version 2.8.0
TLT Version (if relevant)

How do i edit the vocab files? i tried to copy out to my local host but the file is locked.

rvinobha · December 13, 2022, 6:38am

HI @emailmeemilyong

Thanks for your interest in Riva

Please find the below link on Custom Vocabulary Usage

https://docs.nvidia.com/deeplearning/riva/user-guide/docs/tutorials/asr-how-to-improve-recognition-for-specific-words.html?highlight=vocab#custom-vocabulary

Please find the below notebook

github.com

nvidia-riva/tutorials/blob/stable/asr-python-advanced-customize-vocabulary-and-lexicon.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"http://developer.download.nvidia.com/notebooks/dlsw-notebooks/riva_asr_asr-python-advanced-customize-vocabulary-and-lexicon/nvidia_logo.png\" style=\"width: 90px; float: right;\">\n",
    "\n",
    "# How to Customize Riva ASR Vocabulary and Pronunciation with Lexicon Mapping\n",
    "\n",
    "This notebook walks you through the process of customizing Riva ASR vocabulary and lexicon, in order to improve Riva vocabulary coverage and recognition of difficult words, such as acronyms.\n",
    "\n",
    "## Overview\n",
    "\n",
    "The Flashlight decoder, deployed by default in Riva, is a lexicon-based decoder and only emits words that are present in the provided lexicon file. That means, uncommon and new words, such as domain specific terminologies, that are not present in the lexicon file, will have no chance of being generated.\n",
    "\n",
    "On the other hand, the greedy decoder (available as an option during the `riva-build` process with the flag `--decoder_type=greedy`) is not lexicon-based and hence can virtually produce any word or character sequence.\n",
    "\n",
    "### Pre requisite\n",
    "\n",

This file has been truncated. show original

you will find information on Extending the vocabulary

Thanks