AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

SpeechOp: Inference-Time Task Composition for Generative Speech Processing

Justin Lovelace, Rithesh Kumar, Jiaqi Su, Ke Chen, Kilian Q Weinberger, Zeyu Jin

ICLR 2026 · Image generation › Autoregressive and latent generation · approach: Neural text-to-speech synthesis · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add multi-task training (enhancement, separation) to the TTS model (ablation)TTS Baseline · Zero-shot TTSMOS-VS: 3.31 -> 3.67Improved
add multi-task training to the TTS model (ablation)TTS Baseline · Zero-shot TTSMOS-Q: 3.65 -> 3.86Improved
add multi-task training to the TTS model (ablation)TTS Baseline · Zero-shot TTSWER: 3.32 -> 3.57Hurt
add multi-task training to the TTS model (ablation)TTS Baseline · Zero-shot TTSspeaker similarity: .48 -> .53Improved
add multi-task training to the TTS model (ablation)TTS Baseline · Speech editingMOS-Q: 4.18 -> 4.15No clear effect
add implicit task composition with Whisper transcriptsSpeechOp enhancement · Speech enhancementWER: 8.1 -> 2.9Improved
add implicit task composition with Whisper transcriptsSpeechOp enhancement · Speech enhancementPESQ: 2.00 -> 2.05Improved
add speaker personalization to ITC enhancementSpeechOp-ITC · Speech enhancementMCD: 4.85 -> 4.69Improved
use TC-CFG instead of score averaging (TC-Avg) (ablation)SpeechOp enhancement · Speech enhancement (gold transcript)WER: 3.4 -> 2.1Improved
use TC-CFG instead of score averaging (TC-Avg) (ablation)SpeechOp enhancement · Speech enhancement (gold transcript)MCD: 5.24 -> 4.83Improved
use TC-CFG instead of score averaging (TC-Avg) (ablation)SpeechOp enhancement · Speech enhancement (gold transcript)PESQ: 1.88 -> 2.06Improved
use framewise mixing instead of cross-attention for source audio (ablation)SpeechOp-Small · Speech enhancementWER: 9.6 -> 8.8Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy