Channel fusion gives better semantic grounding and QA performance in full-duplex LLM dialogue but is vulnerable to context corruption during interruptions, while cross-attention routing is more robust at the cost of weaker integration.
Emilia: A large-scale, extensive, multilin- gual, and diverse dataset for speech generation
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
dataset 2
citation-polarity summary
verdicts
UNVERDICTED 5roles
dataset 2polarities
use dataset 2representative citing papers
CleanCodec reframes audio tokenization as a selective information bottleneck to encode only perceptually important features at 12.5 tokens per second, outperforming prior codecs in efficiency, speaker similarity, and intelligibility.
The paper introduces target-KL regularization to train audio VAEs at specific bitrates, enabling rate-distortion curves and comparison to discrete audio codecs for improved text-to-sound generation.
Replacing early-reflected speech with time-shifted anechoic clean speech as the training target, combined with a two-stage distortion-perception framework, yields state-of-the-art universal speech enhancement.
Kimi-Audio is an open-source audio foundation model that achieves state-of-the-art results on speech recognition, audio understanding, question answering, and conversation after pre-training on more than 13 million hours of speech, sound, and music data.
citing papers explorer
-
How Should LLMs Listen While Speaking? A Study of User-Stream Routing in Full-Duplex Spoken Dialogue
Channel fusion gives better semantic grounding and QA performance in full-duplex LLM dialogue but is vulnerable to context corruption during interruptions, while cross-attention routing is more robust at the cost of weaker integration.
-
CleanCodec: Efficient and Robust Speech Tokenization via Perceptually Guided Encoding
CleanCodec reframes audio tokenization as a selective information bottleneck to encode only perceptually important features at 12.5 tokens per second, outperforming prior codecs in efficiency, speaker similarity, and intelligibility.
-
Taming Audio VAEs via Target-KL Regularization
The paper introduces target-KL regularization to train audio VAEs at specific bitrates, enabling rate-distortion curves and comparison to discrete audio codecs for improved text-to-sound generation.
-
Rethinking Training Targets, Architectures and Data Quality for Universal Speech Enhancement
Replacing early-reflected speech with time-shifted anechoic clean speech as the training target, combined with a two-stage distortion-perception framework, yields state-of-the-art universal speech enhancement.
-
Kimi-Audio Technical Report
Kimi-Audio is an open-source audio foundation model that achieves state-of-the-art results on speech recognition, audio understanding, question answering, and conversation after pre-training on more than 13 million hours of speech, sound, and music data.