Speaker identification on the SCOTUS corpus

By Jiahong Yuan, Mark Y. Liberman. 2008

676 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Natural Language Processing, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

In Signal Processing

Citations in the corpus

Multimodal Transformer for Unaligned Multimodal Language Sequences (2019, methodology): “, 2019), we first use P2FA (Yuan and Liberman, 2008) to obtain the aligned timesteps (segmented w.”
A3T: Alignment-Aware Acoustic and Text Pretraining for Speech Synthesis and Editing (2022, positive): “We use a forced aligner (Yuan & Liberman, 2008) to pre-process the dataset to get the alignment information, which is shown in Fig.”
Improving Multimodal Fusion with Hierarchical Mutual Information Maximization for Multimodal Sentiment Analysis (2021, methodology): “For visual and acoustic, we use COVAREP (Degot-tex et al., 2014) and P2FA (Yuan and Liberman, 2008), which both are prevalent tool kits for feature extraction and have been regularly employed before.”
Lip Reading in Profile (2017, methodology): “As done in previous works [9], the Penn Forced Aligner [26] is used to align the subtitle to the audio speech; and the force-aligned subtitles are doublechecked against a transcript given by a commercial speech recognition software.”
LRS3-TED: a large-scale dataset for visual speech recognition (2018, neutral): “The subtitles in the YouTube videos are broadcast in sync with the audio only at sentence-level, therefore the Penn Phonetics Lab Forced Aligner (P2FA) [12] is used to obtain a word-level alignment between the subtitle and the audio signal.”
Memory Fusion Network for Multi-view Sequential Learning (2018, methodology): “To get the exact utterance timestamp of each word we perform forced alignment using P2FA (Yuan and Liberman 2008) which allows us to align the three modalities together.”
Controllable Neural Prosody Synthesis (2020, neutral): “We use P2FA [27] for phoneme alignment.”
Lip Reading Sentences in the Wild (2016, methodology): “The Penn Phonetics Lab Forced Aligner [18, 39] is used to force-align the subtitle to the audio signal.”
Montreal Forced Aligner: Trainable Text-Speech Alignment Using Kaldi (2017, methodology): “The distributions of differences are highly rightskewed, as for other forced aligners [8, 26]: 2–5% of tokens have differences of at least 100 msec, while about 90% have differences of less than 50 msec.”
Text-based editing of talking-head video (2019, methodology): “To segment the video’s speech audio into phones (audible realizations of phonemes), we assume we have an accurate text transcript and align it to the audio using P2FA [Rubin et al. 2013; Yuan and Liberman 2008], a phoneme-based alignment tool.”
Multimodal Language Analysis with Recurrent Multistage Fusion (2018, methodology): “Forced alignment is performed using P2FA (Yuan and Liberman, 2008) to obtain the exact utterance times 1Details on feature extraction are in supplementary. of each word.”
Efficient Low-rank Multimodal Fusion With Modality-Specific Factors (2018, methodology): “To reach the same time alignment across modalities, we perform word alignment using P2FA (Yuan and Liberman, 2008) which allows us to align the three modalities at the word granularity.”
Natural speech reveals the semantic maps that tile human cerebral cortex (2016, positive): “The audio of each story was downsampled to 11 kHz and the Penn Phonetics Lab Forced Aligner (P2FA 33 ) was used to automatically align the audio to the transcript.”
Factorized Multimodal Transformer for Multimodal Sequential Learning (2019, methodology): “For BA and F1, we report two numbers: the number on the left side of “/” is calculated based on approach taken by Zadeh et al. (2018b), and the right side is by Tsai et al.”
Found in Translation: Learning Robust Joint Representations by Cyclic Translations Between Modalities (2018, neutral): “Forced alignment is performed using P2FA (Yuan and Liberman 2008) to obtain spoken word utterance times.”