PUBLICATIONS.md

Publications

Here, we list a collection of research articles that utilize the NeMo Toolkit. If you would like to include your paper in this collection, please submit a PR updating this document.

---

Automatic Speech Recognition (ASR)

2021

• Citrinet: Closing the Gap between Non-Autoregressive and Autoregressive End-to-End Models for Automatic Speech Recognition
• SPGISpeech: 5,000 hours of transcribed financial audio for fully formatted end-to-end speech recognition
• CarneliNet: Neural Mixture Model for Automatic Speech Recognition
• CTC Variations Through New WFST Topologies
• A Toolbox for Construction and Analysis of Speech Datasets

2020

• Cross-Language Transfer Learning, Continuous Learning, and Domain Adaptation for End-to-End Automatic Speech Recognition
• Correction of Automatic Speech Recognition with Transformer Sequence-To-Sequence Model
• Improving Noise Robustness of an End-to-End Neural Model for Automatic Speech Recognition

2019

• Jasper: An End-to-End Convolutional Neural Acoustic Model
• QuartzNet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutions

---

Speaker Recognition (SpkR)

2021

• TitaNet: Neural Model for speaker representation with 1D Depth-wise separable convolutions and global context

2020

• SpeakerNet: 1D Depth-wise Separable Convolutional Network for Text-Independent Speaker Recognition and Verification

---

Speech Classification

2021

• MarbleNet: Deep 1D Time-Channel Separable Convolutional Neural Network for Voice Activity Detection

2020

• MatchboxNet - 1D Time-Channel Separable Convolutional Neural Network Architecture for Speech Commands Recognition

---

Natural Language Processing (NLP)

Language Modeling

2021

• BioMegatron: Larger Biomedical Domain Language Model

---

Dialogue State Tracking

2021

• SGD-QA: Fast Schema-Guided Dialogue State Tracking for Unseen Services

---

Text To Speech (TTS)

2021

• TalkNet: Fully-Convolutional Non-Autoregressive Speech Synthesis Model
• TalkNet 2: Non-Autoregressive Depth-Wise Separable Convolutional Model for Speech Synthesis with Explicit Pitch and Duration Prediction
• Hi-Fi Multi-Speaker English TTS Dataset
• Mixer-TTS: non-autoregressive, fast and compact text-to-speech model conditioned on language model embeddings

---

(Inverse) Text Normalization

2021

• NeMo Inverse Text Normalization: From Development to Production
• A Unified Transformer-based Framework for Duplex Text Normalization

---