speech-4

Acoustic Modelling

3260 papers • 126 benchmarks • 313 datasets

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

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in acoustic-modelling-4

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find acoustic-modelling-4 models and implementations

Datasets

No datasets available.

Subtasks

No subtasks available.

Most implemented papers

Phonemic Transcription of Low-Resource Tonal Languages

Graham Neubig, Trevor Cohn, Oliver Adams, Alexis Michaud•Tue Dec 05 2017

The use of a neural network architecture with the connectionist temporal classification loss function for phonemic and tonal transcription in a language documentation setting is explored and the method's promise in improving efficiency, minimizing typographical errors, and maintaining the transcription's faithfulness to the acoustic signal is shown.

16

Content

0

Paper Graph

End-to-end attention-based large vocabulary speech recognition

Dmitriy Serdyuk, Yoshua Bengio, Dzmitry Bahdanau, J. Chorowski, Philemon Brakel•Mon Aug 17 2015

This work investigates an alternative method for sequence modelling based on an attention mechanism that allows a Recurrent Neural Network (RNN) to learn alignments between sequences of input frames and output labels.

1193 0

Paper Graph

GIBBONFINDR: An R package for the detection and classification of acoustic signals

H. Klinck, D. Clink•Wed Jun 05 2019

The new, open-source R package GIBBONFINDR is described which has functions for detection, classification and visualization of acoustic signals using a variety of readily available machine learning algorithms in the R programming environment.

7 0

Paper Graph

Acoustic Model Adaptation from Raw Waveforms with Sincnet

S. Renals, Ondrej Klejch, P. Bell, Joachim Fainberg, Erfan Loweimi•Sun Sep 29 2019

It is shown that the parameterisation of the SincNet layer is well suited for adaptation in practice: it can efficiently adapt with a very small number of parameters, producing error rates comparable to techniques using orders of magnitude more parameters.

15 0

Paper Graph

Multilingual Bottleneck Features for Improving ASR Performance of Code-Switched Speech in Under-Resourced Languages

Trideba Padhi, A. Biswas, F. D. Wet, E. V. D. Westhuizen, T. Niesler•Fri Oct 30 2020

This work explores the benefits of using multilingual bottleneck features (mBNF) in acoustic modelling for the automatic speech recognition of code-switched speech in African languages and shows that the inclusion of the mBNF features leads to clear performance improvements over a baseline trained without them.

4 0

Paper Graph

Matcha-TTS: A Fast TTS Architecture with Conditional Flow Matching

G. Henter, J. Beskow, Shivam Mehta, Éva Székely, Ruibo Tu•Tue Sep 05 2023

Matcha-TTS, a new encoder-decoder architecture for speedy TTS acoustic modelling, trained using optimal-transport conditional flow matching (OT-CFM), is introduced, an ODE-based decoder capable of high output quality in fewer synthesis steps than models trained using score matching.

185 0

Paper Graph

Adding a benchmark result helps the community track progress.