Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

"speech": models, code, and papers

Integrate Document Ranking Information into Confidence Measure Calculation for Spoken Term Detection

Sep 10, 2015
Quan Liu, Wu Guo, Zhen-Hua Ling

This paper proposes an algorithm to improve the calculation of confidence measure for spoken term detection (STD). Given an input query term, the algorithm first calculates a measurement named document ranking weight for each document in the speech database to reflect its relevance with the query term by summing all the confidence measures of the hypothesized term occurrences in this document. The confidence measure of each term occurrence is then re-estimated through linear interpolation with the calculated document ranking weight to improve its reliability by integrating document-level information. Experiments are conducted on three standard STD tasks for Tamil, Vietnamese and English respectively. The experimental results all demonstrate that the proposed algorithm achieves consistent improvements over the state-of-the-art method for confidence measure calculation. Furthermore, this algorithm is still effective even if a high accuracy speech recognizer is not available, which makes it applicable for the languages with limited speech resources.

* 4 pages 

  Access Paper or Ask Questions

Design of a novel Korean learning application for efficient pronunciation correction

May 04, 2022
Minjong Cheon, Minseon Kim, Hanseon Joo

The Korean wave, which denotes the global popularity of South Korea's cultural economy, contributes to the increasing demand for the Korean language. However, as there does not exist any application for foreigners to learn Korean, this paper suggested a design of a novel Korean learning application. Speech recognition, speech-to-text, and speech-to-waveform are the three key systems in the proposed system. The Google API and the librosa library will transform the user's voice into a sentence and MFCC. The software will then display the user's phrase and answer, with mispronounced elements highlighted in red, allowing users to more easily recognize the incorrect parts of their pronunciation. Furthermore, the Siamese network might utilize those translated spectrograms to provide a similarity score, which could subsequently be used to offer feedback to the user. Despite the fact that we were unable to collect sufficient foreigner data for this research, it is notable that we presented a novel Korean pronunciation correction method for foreigners.

  Access Paper or Ask Questions

Multimodal analysis of the predictability of hand-gesture properties

Aug 12, 2021
Taras Kucherenko, Rajmund Nagy, Michael Neff, Hedvig Kjellström, Gustav Eje Henter

Embodied conversational agents benefit from being able to accompany their speech with gestures. Although many data-driven approaches to gesture generation have been proposed in recent years, it is still unclear whether such systems can consistently generate gestures that convey meaning. We investigate which gesture properties (phase, category, and semantics) can be predicted from speech text and/or audio using contemporary deep learning. In extensive experiments, we show that gesture properties related to gesture meaning (semantics and category) are predictable from text features (time-aligned BERT embeddings) alone, but not from prosodic audio features, while rhythm-related gesture properties (phase) on the other hand can be predicted from either audio, text (with word-level timing information), or both. These results are encouraging as they indicate that it is possible to equip an embodied agent with content-wise meaningful co-speech gestures using a machine-learning model.

* 10 pages, 7 figures 

  Access Paper or Ask Questions

Naturalization of Text by the Insertion of Pauses and Filler Words

Nov 07, 2020
Richa Sharma, Parth Vipul Shah, Ashwini M. Joshi

In this article, we introduce a set of methods to naturalize text based on natural human speech. Voice-based interactions provide a natural way of interfacing with electronic systems and are seeing a widespread adaptation of late. These computerized voices can be naturalized to some degree by inserting pauses and filler words at appropriate positions. The first proposed text transformation method uses the frequency of bigrams in the training data to make appropriate insertions in the input sentence. It uses a probability distribution to choose the insertions from a set of all possible insertions. This method is fast and can be included before a Text-To-Speech module. The second method uses a Recurrent Neural Network to predict the next word to be inserted. It confirms the insertions given by the bigram method. Additionally, the degree of naturalization can be controlled in both these methods. On the conduction of a blind survey, we conclude that the output of these text transformation methods is comparable to natural speech.

* Keywords: Text transformation, natural speech, bigram, RNN, filler words 

  Access Paper or Ask Questions

Towards non-toxic landscapes: Automatic toxic comment detection using DNN

Nov 19, 2019
Ashwin Geet D'Sa, Irina Illina, Dominique Fohr

The spectacular expansion of the Internet led to the development of a new research problem in the natural language processing field: automatic toxic comment detection, since many countries prohibit hate speech in public media. There is no clear and formal definition of hate, offensive, toxic and abusive speeches. In this article, we put all these terms under the "umbrella" of toxic speech. The contribution of this paper is the design of binary classification and regression-based approaches aiming to predict whether a comment is toxic or not. We compare different unsupervised word representations and different DNN classifiers. Moreover, we study the robustness of the proposed approaches to adversarial attacks by adding one (healthy or toxic) word. We evaluate the proposed methodology on the English Wikipedia Detox corpus. Our experiments show that using BERT fine-tuning outperforms feature-based BERT, Mikolov's word embedding or fastText representations with different DNN classifiers.

  Access Paper or Ask Questions

Prosodic Alignment for off-screen automatic dubbing

Apr 06, 2022
Yogesh Virkar, Marcello Federico, Robert Enyedi, Roberto Barra-Chicote

The goal of automatic dubbing is to perform speech-to-speech translation while achieving audiovisual coherence. This entails isochrony, i.e., translating the original speech by also matching its prosodic structure into phrases and pauses, especially when the speaker's mouth is visible. In previous work, we introduced a prosodic alignment model to address isochrone or on-screen dubbing. In this work, we extend the prosodic alignment model to also address off-screen dubbing that requires less stringent synchronization constraints. We conduct experiments on four dubbing directions - English to French, Italian, German and Spanish - on a publicly available collection of TED Talks and on publicly available YouTube videos. Empirical results show that compared to our previous work the extended prosodic alignment model provides significantly better subjective viewing experience on videos in which on-screen and off-screen automatic dubbing is applied for sentences with speakers mouth visible and not visible, respectively.

* 5 pages, 2 figures, 3 tables, Submitted to Interspeech 2022 

  Access Paper or Ask Questions

Statistical Testing on ASR Performance via Blockwise Bootstrap

Dec 19, 2019
Zhe Liu

A common question being raised in automatic speech recognition (ASR) evaluations is how reliable is an observed word error rate (WER) improvement comparing two ASR systems, where statistical hypothesis testing and confidence intervals can be utilized to tell whether this improvement is real or only due to random chance. The bootstrap resampling method has been popular for such significance analysis which is intuitive and easy to use. However, this method fails in dealing with dependent data, which is prevalent in speech world - for example, ASR performance on utterances from the same speaker could be correlated. In this paper we present blockwise bootstrap approach - by dividing evaluation utterances into nonoverlapping blocks, this method resamples these blocks instead of original data. We show that the resulting variance estimator of absolute WER difference of two ASR systems is consistent under mild conditions. We also demonstrate the validity of blockwise bootstrap method on both synthetic and real-world speech data.

* 8 pages, 2 figures 

  Access Paper or Ask Questions

Synchronising audio and ultrasound by learning cross-modal embeddings

Jul 01, 2019
Aciel Eshky, Manuel Sam Ribeiro, Korin Richmond, Steve Renals

Audiovisual synchronisation is the task of determining the time offset between speech audio and a video recording of the articulators. In child speech therapy, audio and ultrasound videos of the tongue are captured using instruments which rely on hardware to synchronise the two modalities at recording time. Hardware synchronisation can fail in practice, and no mechanism exists to synchronise the signals post hoc. To address this problem, we employ a two-stream neural network which exploits the correlation between the two modalities to find the offset. We train our model on recordings from 69 speakers, and show that it correctly synchronises 82.9% of test utterances from unseen therapy sessions and unseen speakers, thus considerably reducing the number of utterances to be manually synchronised. An analysis of model performance on the test utterances shows that directed phone articulations are more difficult to automatically synchronise compared to utterances containing natural variation in speech such as words, sentences, or conversations.

* 5 pages, 1 figure, 4 tables; accepted to Interspeech 2019: the 20th Annual Conference of the International Speech Communication Association (ISCA) 

  Access Paper or Ask Questions

On-the-Fly Aligned Data Augmentation for Sequence-to-Sequence ASR

Apr 03, 2021
Tsz Kin Lam, Mayumi Ohta, Shigehiko Schamoni, Stefan Riezler

We propose an on-the-fly data augmentation method for automatic speech recognition (ASR) that uses alignment information to generate effective training samples. Our method, called Aligned Data Augmentation (ADA) for ASR, replaces transcribed tokens and the speech representations in an aligned manner to generate previously unseen training pairs. The speech representations are sampled from an audio dictionary that has been extracted from the training corpus and inject speaker variations into the training examples. The transcribed tokens are either predicted by a language model such that the augmented data pairs are semantically close to the original data, or randomly sampled. Both strategies result in training pairs that improve robustness in ASR training. Our experiments on a Seq-to-Seq architecture show that ADA can be applied on top of SpecAugment, and achieves about 9-23% and 4-15% relative improvements in WER over SpecAugment alone on LibriSpeech 100h and LibriSpeech 960h test datasets, respectively.

* Submitted to INTERSPEECH 2021 

  Access Paper or Ask Questions

Libri-Light: A Benchmark for ASR with Limited or No Supervision

Dec 17, 2019
Jacob Kahn, Morgane Rivière, Weiyi Zheng, Evgeny Kharitonov, Qiantong Xu, Pierre-Emmanuel Mazaré, Julien Karadayi, Vitaliy Liptchinsky, Ronan Collobert, Christian Fuegen, Tatiana Likhomanenko, Gabriel Synnaeve, Armand Joulin, Abdelrahman Mohamed, Emmanuel Dupoux

We introduce a new collection of spoken English audio suitable for training speech recognition systems under limited or no supervision. It is derived from open-source audio books from the LibriVox project. It contains over 60K hours of audio, which is, to our knowledge, the largest freely-available corpus of speech. The audio has been segmented using voice activity detection and is tagged with SNR, speaker ID and genre descriptions. Additionally, we provide baseline systems and evaluation metrics working under three settings: (1) the zero resource/unsupervised setting (ABX), (2) the semi-supervised setting (PER, CER) and (3) the distant supervision setting (WER). Settings (2) and (3) use limited textual resources (10 minutes to 10 hours) aligned with the speech. Setting (3) uses large amounts of unaligned text. They are evaluated on the standard LibriSpeech dev and test sets for comparison with the supervised state-of-the-art.

  Access Paper or Ask Questions