Directed Speech Separation For Automatic Speech Recognition Of Long Form Conversational Speech
2021 · Rohit Paturi, Sundararajan Srinivasan, Katrin Kirchhoff, et al.
Abstract
Many of the recent advances in speech separation are primarily aimed at synthetic mixtures of short audio utterances with high degrees of overlap. Most of these approaches need an additional stitching step to stitch the separated speech chunks for long form audio. Since most of the approaches involve Permutation Invariant training (PIT), the order of separated speech chunks is nondeterministic and leads to difficulty in accurately stitching homogenous speaker chunks for downstream tasks like Automatic Speech Recognition (ASR). Also, most of these models are trained with synthetic mixtures and do not generalize to real conversational data. In this paper, we propose a speaker conditioned separator trained on speaker embeddings extracted directly from the mixed signal using an over-clustering based approach. This model naturally regulates the order of the separated chunks without the need for an additional stitching step. We also introduce a data sampling strategy with real and synthetic
Authors
(none)
Tags
Stats
Related papers
- Separating Long-form Speech With Group-wise Permutation Invariant Training (2021)4.52
- Continuous Speech Separation Using Speaker Inventory For Long Multi-talker Recording (2020)7.50
- Transcription-free Fine-tuning Of Speech Separation Models For Noisy And Reverberant Multi-speaker Automatic Speech Recognition (2024)3.58
- Individualized Conditioning And Negative Distances For Speaker Separation (2022)2.26
- Permutation Invariant Training Of Deep Models For Speaker-independent Multi-talker Speech Separation (2016)0.00
- Discriminative Learning For Monaural Speech Separation Using Deep Embedding Features (2019)8.60
- A Sidecar Separator Can Convert A Single-talker Speech Recognition System To A Multi-talker One (2023)9.03
- Separate And Reconstruct: Asymmetric Encoder-decoder For Speech Separation (2024)0.00