Grapheme-to-Phoneme Conversion

16 papers with code • 0 benchmarks • 0 datasets

This task has no description! Would you like to contribute one?

Most implemented papers

Deep Voice: Real-time Neural Text-to-Speech

NVIDIA/nv-wavenet ICML 2017

We present Deep Voice, a production-quality text-to-speech system constructed entirely from deep neural networks.

Applying the Transformer to Character-level Transduction

shijie-wu/neural-transducer EACL 2021

The transformer has been shown to outperform recurrent neural network-based sequence-to-sequence models in various word-level NLP tasks.

Jointly Learning to Align and Convert Graphemes to Phonemes with Neural Attention Models

shtoshni92/g2p 20 Oct 2016

We propose an attention-enabled encoder-decoder model for the problem of grapheme-to-phoneme conversion.

Massively Multilingual Neural Grapheme-to-Phoneme Conversion

bpopeters/mg2p WS 2017

Grapheme-to-phoneme conversion (g2p) is necessary for text-to-speech and automatic speech recognition systems.

g2pM: A Neural Grapheme-to-Phoneme Conversion Package for Mandarin Chinese Based on a New Open Benchmark Dataset

kakaobrain/g2pM 7 Apr 2020

Conversion of Chinese graphemes to phonemes (G2P) is an essential component in Mandarin Chinese Text-To-Speech (TTS) systems.

Transformer based Grapheme-to-Phoneme Conversion

as-ideas/deepphonemizer arXiv preprint 2019

The transformer network architecture is completely based on attention mechanisms, and it outperforms sequence-to-sequence models in neural machine translation without recurrent and convolutional layers.

Supervised Grapheme-to-Phoneme Conversion of Orthographic Schwas in Hindi and Punjabi

aryamanarora/schwa-deletion ACL 2020

Hindi grapheme-to-phoneme (G2P) conversion is mostly trivial, with one exception: whether a schwa represented in the orthography is pronounced or unpronounced (deleted).

Smoothing and Shrinking the Sparse Seq2Seq Search Space

deep-spin/S7 NAACL 2021

Current sequence-to-sequence models are trained to minimize cross-entropy and use softmax to compute the locally normalized probabilities over target sequences.

On Biasing Transformer Attention Towards Monotonicity

ZurichNLP/monotonicity_loss NAACL 2021

Many sequence-to-sequence tasks in natural language processing are roughly monotonic in the alignment between source and target sequence, and previous work has facilitated or enforced learning of monotonic attention behavior via specialized attention functions or pretraining.