• Publications
  • Influence
Building Speech Recognition Systems for Language Documentation: The CoEDL Endangered Language Pipeline and Inference System (ELPIS)
TLDR
This paper describes the development of Elpis, a pipeline which language documentation workers with minimal computational experience can use to build their own speech recognition models, resulting in models being built for 16 languages from the Asia-Pacific region. Expand
  • 19
  • 2
  • PDF
Text Normalization Infrastructure that Scales to Hundreds of Language Varieties
TLDR
We describe the automated multi-language text normalization infrastructure that prepares textual data to train language models used in Google’s keyboards and speech recognition systems, across hundreds of language varieties. Expand
  • 11
  • 1
  • PDF
Predicting Pronunciations with Syllabification and Stress with Recurrent Neural Networks
TLDR
We train recurrent neural network (RNN) based models to predict the phoneme sequence, syllabification and stress pattern for such pronunciations making them usable for TTS. Expand
  • 6
  • 1
An Expanded Taxonomy of Semiotic Classes for Text Normalization
TLDR
We describe an expanded taxonomy of semiotic classes for text normalization, building upon the work in [1]. Expand
  • 8
  • 1
Automatic Keyboard Layout Design for Low-Resource Latin-Script Languages
TLDR
We present our approach to automatically designing and implementing keyboard layouts on mobile devices for typing low-resource languages written in the Latin script. Expand
  • 4
  • 1
  • PDF
Writing Across the World's Languages: Deep Internationalization for Gboard, the Google Keyboard
TLDR
We describe how and why we have been adding support for hundreds of language varieties from around the globe, and we describe the trends we see. Expand
  • 8
  • PDF
Unified Verbalization for Speech Recognition & Synthesis Across Languages
TLDR
We describe a new approach to converting written tokens to their spoken form, which can be shared by automatic speech recognition (ASR) and text-to-speech synthesis (TTS) systems. Expand
  • 4
  • PDF
Building Large-Vocabulary ASR Systems for Languages Without Any Audio Training Data
TLDR
We show that it is possible to build large-vocabulary ASR systems for languages with no audio training data at all, as long as a high-resource language that is sufficiently phonologically similar can be identified. Expand
  • 5
  • PDF
Language ID in the Wild: Unexpected Challenges on the Path to a Thousand-Language Web Text Corpus
TLDR
Large text corpora are increasingly important for a wide variety of Natural Language Processing (NLP) tasks, and automatic language identification (LangID) is a core technology needed to collect such datasets in a multilingual context. Expand
  • 3
  • PDF
...
1
2
...