Pith. sign in

REVIEW 1 cited by

Serialized Output Training by Learned Dominance

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2407.03966 v1 pith:SHLGX3FT submitted 2024-07-04 cs.SD cs.AIeess.AS

classification cs.SDcs.AIeess.AS
keywords speechcomponentsoutputtrainingdominancefactorsfifomodule
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Serialized Output Training (SOT) has showcased state-of-the-art performance in multi-talker speech recognition by sequentially decoding the speech of individual speakers. To address the challenging label-permutation issue, prior methods have relied on either the Permutation Invariant Training (PIT) or the time-based First-In-First-Out (FIFO) rule. This study presents a model-based serialization strategy that incorporates an auxiliary module into the Attention Encoder-Decoder architecture, autonomously identifying the crucial factors to order the output sequence of the speech components in multi-talker speech. Experiments conducted on the LibriSpeech and LibriMix databases reveal that our approach significantly outperforms the PIT and FIFO baselines in both 2-mix and 3-mix scenarios. Further analysis shows that the serialization module identifies dominant speech components in a mixture by factors including loudness and gender, and orders speech components based on the dominance score.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Speaker Targeting via Self-Speaker Adaptation for Multi-talker ASR

    eess.AS 2025-06 conditional novelty 6.0 of 10

    A speaker-activity mask injected into an ASR encoder lets one model instance transcribe each talker in overlapped speech without speaker embeddings.

Pith tools