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

Speaker Recognition in the Wild

May 05, 2022
Neeraj Chhimwal, Anirudh Gupta, Rishabh Gaur, Harveen Singh Chadha, Priyanshi Shah, Ankur Dhuriya, Vivek Raghavan

In this paper, we propose a pipeline to find the number of speakers, as well as audios belonging to each of these now identified speakers in a source of audio data where number of speakers or speaker labels are not known a priori. We used this approach as a part of our Data Preparation pipeline for Speech Recognition in Indic Languages (https://github.com/Open-Speech-EkStep/vakyansh-wav2vec2-experimentation). To understand and evaluate the accuracy of our proposed pipeline, we introduce two metrics: Cluster Purity, and Cluster Uniqueness. Cluster Purity quantifies how "pure" a cluster is. Cluster Uniqueness, on the other hand, quantifies what percentage of clusters belong only to a single dominant speaker. We discuss more on these metrics in section \ref{sec:metrics}. Since we develop this utility to aid us in identifying data based on speaker IDs before training an Automatic Speech Recognition (ASR) model, and since most of this data takes considerable effort to scrape, we also conclude that 98\% of data gets mapped to the top 80\% of clusters (computed by removing any clusters with less than a fixed number of utterances -- we do this to get rid of some very small clusters and use this threshold as 30), in the test set chosen.

* This paper was submitted to Interspeech 2022 

  Access Paper or Ask Questions

One TTS Alignment To Rule Them All

Aug 23, 2021
Rohan Badlani, Adrian Łancucki, Kevin J. Shih, Rafael Valle, Wei Ping, Bryan Catanzaro

Speech-to-text alignment is a critical component of neural textto-speech (TTS) models. Autoregressive TTS models typically use an attention mechanism to learn these alignments on-line. However, these alignments tend to be brittle and often fail to generalize to long utterances and out-of-domain text, leading to missing or repeating words. Most non-autoregressive endto-end TTS models rely on durations extracted from external sources. In this paper we leverage the alignment mechanism proposed in RAD-TTS as a generic alignment learning framework, easily applicable to a variety of neural TTS models. The framework combines forward-sum algorithm, the Viterbi algorithm, and a simple and efficient static prior. In our experiments, the alignment learning framework improves all tested TTS architectures, both autoregressive (Flowtron, Tacotron 2) and non-autoregressive (FastPitch, FastSpeech 2, RAD-TTS). Specifically, it improves alignment convergence speed of existing attention-based mechanisms, simplifies the training pipeline, and makes the models more robust to errors on long utterances. Most importantly, the framework improves the perceived speech synthesis quality, as judged by human evaluators.


  Access Paper or Ask Questions

Word Error Rate Estimation Without ASR Output: e-WER2

Aug 08, 2020
Ahmed Ali, Steve Renals

Measuring the performance of automatic speech recognition (ASR) systems requires manually transcribed data in order to compute the word error rate (WER), which is often time-consuming and expensive. In this paper, we continue our effort in estimating WER using acoustic, lexical and phonotactic features. Our novel approach to estimate the WER uses a multistream end-to-end architecture. We report results for systems using internal speech decoder features (glass-box), systems without speech decoder features (black-box), and for systems without having access to the ASR system (no-box). The no-box system learns joint acoustic-lexical representation from phoneme recognition results along with MFCC acoustic features to estimate WER. Considering WER per sentence, our no-box system achieves 0.56 Pearson correlation with the reference evaluation and 0.24 root mean square error (RMSE) across 1,400 sentences. The estimated overall WER by e-WER2 is 30.9% for a three hours test set, while the WER computed using the reference transcriptions was 28.5%.


  Access Paper or Ask Questions

A Deep Learning Approach for Similar Languages, Varieties and Dialects

Jan 02, 2019
Vidya Prasad K, Akarsh S, Vinayakumar R, Soman KP

Deep learning mechanisms are prevailing approaches in recent days for the various tasks in natural language processing, speech recognition, image processing and many others. To leverage this we use deep learning based mechanism specifically Bidirectional- Long Short-Term Memory (B-LSTM) for the task of dialectic identification in Arabic and German broadcast speech and Long Short-Term Memory (LSTM) for discriminating between similar Languages. Two unique B-LSTM models are created using the Large-vocabulary Continuous Speech Recognition (LVCSR) based lexical features and a fixed length of 400 per utterance bottleneck features generated by i-vector framework. These models were evaluated on the VarDial 2017 datasets for the tasks Arabic, German dialect identification with dialects of Egyptian, Gulf, Levantine, North African, and MSA for Arabic and Basel, Bern, Lucerne, and Zurich for German. Also for the task of Discriminating between Similar Languages like Bosnian, Croatian and Serbian. The B-LSTM model showed accuracy of 0.246 on lexical features and accuracy of 0.577 bottleneck features of i-Vector framework.

* 17 pages 

  Access Paper or Ask Questions

Investigating context features hidden in End-to-End TTS

Nov 04, 2018
Kohki Mametani, Tsuneo Kato, Seiichi Yamamoto

Recent studies have introduced end-to-end TTS, which integrates the production of context and acoustic features in statistical parametric speech synthesis. As a result, a single neural network replaced laborious feature engineering with automated feature learning. However, little is known about what types of context information end-to-end TTS extracts from text input before synthesizing speech, and the previous knowledge about context features is barely utilized. In this work, we first point out the model similarity between end-to-end TTS and parametric TTS. Based on the similarity, we evaluate the quality of encoder outputs from an end-to-end TTS system against eight criteria that are derived from a standard set of context information used in parametric TTS. We conduct experiments using an evaluation procedure that has been newly developed in the machine learning literature for quantitative analysis of neural representations, while adapting it to the TTS domain. Experimental results show that the encoder outputs reflect both linguistic and phonetic contexts, such as vowel reduction at phoneme level, lexical stress at syllable level, and part-of-speech at word level, possibly due to the joint optimization of context and acoustic features.

* Submitted to ICASSP2019 

  Access Paper or Ask Questions

TED-LIUM 3: twice as much data and corpus repartition for experiments on speaker adaptation

Jul 03, 2018
François Hernandez, Vincent Nguyen, Sahar Ghannay, Natalia Tomashenko, Yannick Estève

In this paper, we present TED-LIUM release 3 corpus dedicated to speech recognition in English, that multiplies by more than two the available data to train acoustic models in comparison with TED-LIUM 2. We present the recent development on Automatic Speech Recognition (ASR) systems in comparison with the two previous releases of the TED-LIUM Corpus from 2012 and 2014. We demonstrate that, passing from 207 to 452 hours of transcribed speech training data is really more useful for end-to-end ASR systems than for HMM-based state-of-the-art ones, even if the HMM-based ASR system still outperforms end-to-end ASR system when the size of audio training data is 452 hours, with respectively a Word Error Rate (WER) of 6.6% and 13.7%. Last, we propose two repartitions of the TED-LIUM release 3 corpus: the legacy one that is the same as the one existing in release 2, and a new one, calibrated and designed to make experiments on speaker adaptation. Like the two first releases, TED-LIUM 3 corpus will be freely available for the research community.

* Submitted to SPECOM 2018, 20th International Conference on Speech and Computer; TED-LIUM 3 corpus available on https://lium.univ-lemans.fr/en/ted-lium3/ 

  Access Paper or Ask Questions

Complex spectrogram enhancement by convolutional neural network with multi-metrics learning

Sep 09, 2017
Szu-Wei Fu, Ting-yao Hu, Yu Tsao, Xugang Lu

This paper aims to address two issues existing in the current speech enhancement methods: 1) the difficulty of phase estimations; 2) a single objective function cannot consider multiple metrics simultaneously. To solve the first problem, we propose a novel convolutional neural network (CNN) model for complex spectrogram enhancement, namely estimating clean real and imaginary (RI) spectrograms from noisy ones. The reconstructed RI spectrograms are directly used to synthesize enhanced speech waveforms. In addition, since log-power spectrogram (LPS) can be represented as a function of RI spectrograms, its reconstruction is also considered as another target. Thus a unified objective function, which combines these two targets (reconstruction of RI spectrograms and LPS), is equivalent to simultaneously optimizing two commonly used objective metrics: segmental signal-to-noise ratio (SSNR) and logspectral distortion (LSD). Therefore, the learning process is called multi-metrics learning (MML). Experimental results confirm the effectiveness of the proposed CNN with RI spectrograms and MML in terms of improved standardized evaluation metrics on a speech enhancement task.


  Access Paper or Ask Questions

Pay Better Attention to Attention: Head Selection in Multilingual and Multi-Domain Sequence Modeling

Jun 21, 2021
Hongyu Gong, Yun Tang, Juan Pino, Xian Li

Multi-head attention has each of the attention heads collect salient information from different parts of an input sequence, making it a powerful mechanism for sequence modeling. Multilingual and multi-domain learning are common scenarios for sequence modeling, where the key challenge is to maximize positive transfer and mitigate negative transfer across languages and domains. In this paper, we find that non-selective attention sharing is sub-optimal for achieving good generalization across all languages and domains. We further propose attention sharing strategies to facilitate parameter sharing and specialization in multilingual and multi-domain sequence modeling. Our approach automatically learns shared and specialized attention heads for different languages and domains to mitigate their interference. Evaluated in various tasks including speech recognition, text-to-text and speech-to-text translation, the proposed attention sharing strategies consistently bring gains to sequence models built upon multi-head attention. For speech-to-text translation, our approach yields an average of $+2.0$ BLEU over $13$ language directions in multilingual setting and $+2.0$ BLEU over $3$ domains in multi-domain setting.


  Access Paper or Ask Questions

Deep Feature CycleGANs: Speaker Identity Preserving Non-parallel Microphone-Telephone Domain Adaptation for Speaker Verification

Apr 03, 2021
Saurabh Kataria, Jesús Villalba, Piotr Żelasko, Laureano Moro-Velázquez, Najim Dehak

With the increase in the availability of speech from varied domains, it is imperative to use such out-of-domain data to improve existing speech systems. Domain adaptation is a prominent pre-processing approach for this. We investigate it for adapt microphone speech to the telephone domain. Specifically, we explore CycleGAN-based unpaired translation of microphone data to improve the x-vector/speaker embedding network for Telephony Speaker Verification. We first demonstrate the efficacy of this on real challenging data and then, to improve further, we modify the CycleGAN formulation to make the adaptation task-specific. We modify CycleGAN's identity loss, cycle-consistency loss, and adversarial loss to operate in the deep feature space. Deep features of a signal are extracted from an auxiliary (speaker embedding) network and, hence, preserves speaker identity. Our 3D convolution-based Deep Feature Discriminators (DFD) show relative improvements of 5-10% in terms of equal error rate. To dive deeper, we study a challenging scenario of pooling (adapted) microphone and telephone data with data augmentations and telephone codecs. Finally, we highlight the sensitivity of CycleGAN hyper-parameters and introduce a parameter called probability of adaptation.


  Access Paper or Ask Questions

Gram-CTC: Automatic Unit Selection and Target Decomposition for Sequence Labelling

Aug 12, 2017
Hairong Liu, Zhenyao Zhu, Xiangang Li, Sanjeev Satheesh

Most existing sequence labelling models rely on a fixed decomposition of a target sequence into a sequence of basic units. These methods suffer from two major drawbacks: 1) the set of basic units is fixed, such as the set of words, characters or phonemes in speech recognition, and 2) the decomposition of target sequences is fixed. These drawbacks usually result in sub-optimal performance of modeling sequences. In this pa- per, we extend the popular CTC loss criterion to alleviate these limitations, and propose a new loss function called Gram-CTC. While preserving the advantages of CTC, Gram-CTC automatically learns the best set of basic units (grams), as well as the most suitable decomposition of tar- get sequences. Unlike CTC, Gram-CTC allows the model to output variable number of characters at each time step, which enables the model to capture longer term dependency and improves the computational efficiency. We demonstrate that the proposed Gram-CTC improves CTC in terms of both performance and efficiency on the large vocabulary speech recognition task at multiple scales of data, and that with Gram-CTC we can outperform the state-of-the-art on a standard speech benchmark.

* Published at ICML 2017 

  Access Paper or Ask Questions

<<
360
361
362
363
364
365
366
367
368
369
370
371
372
>>