跳到正文
研踪PaperTrail

论文库

检索计算机领域论文，按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+0新增论文
+0其中 AI 相关
+2已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-15 00:32 · 全站最近入库：2026-11-14 15:09各信源进度 →

统计口径与数据新鲜度

统计 2026-11-14 01:20 — 2026-11-15 01:20（北京时间） · 点击数字查看对应简报；AI 相关包含在新增论文内。

最近完成检查：2026-11-15 00:32最近发现新增版本：2026-11-14 15:09库内论文最新更新时间：2026-11-13 02:59

每分钟刷新已保存的数据。空检查不代表采集异常；新收录可能包含历史论文。AI 导读按新增版本计数。

23,348 已收录论文19,580 AI 相关6,729 有修订112 HF 社区发现

浏览论文库

研究日报 / 周报 →按信源浏览 →

时间范围跟随排序：版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常，查看状态 →

找到 139 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-15 00:30（北京时间）
新论文 · v1arXiv:2611.16933

Kezek: Splitting Mono Two-Person Conversations into Per-Speaker Tracks for Full-Duplex Dialogue Data

Bagzhan Karl, Alizhan Karl

原文摘要

Full-duplex spoken dialogue models learn turn-taking, overlaps and backchannels from conversations recorded with one channel per speaker, but most available conversational audio is a single mono mix. We present Kezek, a model that turns a mono recording of exactly two people into one track per person with the original length and timing; overlaps and short replies ("yeah", "mm-hm") stay on the track of the person who said them. Kezek decides who speaks when and who is who once for the whole recording (stage 1: a two-voice activity model guided by a speaker fingerprint trained to be insensitive to phase), and a conditioned time-domain separator only carries out that decision (stage 2), with outputs that always sum to the input. Standard separation metrics miss the failures that matter for dialogue data, so we report short-reply retention, leakage onto the other person's solo, channel swaps and coverage. Against the open pipeline pyannote -> DialogueSidon on the same recordings, Kezek keeps more short replies (0.83 vs. 0.69 on Open Yap test; 0.80 vs. 0.71 over the other's speech on unseen telephone speech from CallHome English), mostly because it returns the whole recording while the external pipeline drops part of it (on the audio both return, reply retention is about the same: 0.859 vs. 0.865 on Open Yap); it also leaks less, swaps channels less often, and runs 4-5 times faster (about 230 hours of audio per GPU-hour). We also report negative results and an ideal-mask ceiling that locate the remaining gap: quiet replies under a louder voice. Code and weights are released under permissive licences.

eess.AS · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.16787

MESH: Multi-receiver Evaluation of Spatial Hearing across Egocentric and Allocentric Views

Wen Wen, Bohan Li, KwanHo Tam, Yu Xi, Jing Peng, Shuai Wang, Kai Yu

原文摘要

Spatial hearing in multi-receiver environments requires identifying who speaks, where they are, and how they move. Existing benchmarks evaluate spatial prediction, transcription, tracking, and distributed speech processing, but offer limited evaluation of whether speech signal remains associated with the correct physical source across receivers and motion. We introduce MESH, which evaluates this association through egocentric and allocentric views of shared acoustic scenes. The egocentric view uses local recordings and peer metadata, while the allocentric view combines synchronized recordings from multiple receivers in a shared 3D world frame. MESH contains nearly 40,000 questions across 13 static and dynamic localization and spatial transcription tasks. We evaluate eleven audio-model configurations and a language-model agent using specialist tools. The best static distance-estimation and dynamic direction-tracking accuracies are only 21.90% and 41.54%, respectively, while the lowest word error rates across four transcription tasks range from 59.15% to 73.10%. Ablation studies and additional experiments reveal that existing models are not yet able to effectively leverage peer information and struggle to handle multi-speaker and dynamic scenarios. These findings highlight the gap between access to distributed acoustic information and its effective use for spatial speech understanding. We release the dataset and evaluation code to support further research on collaborative and distributed spatial hearing.

eess.AS · cs.SD

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.16718

MusicRWKV: A Multimodal RWKV Framework for Music Representation Learning

Terry Tan

原文摘要

Recent music foundation models have achieved remarkable progress in music representation learning, yet most existing approaches rely on Transformer architectures whose quadratic attention complexity introduces substantial computational and memory overhead when modeling long-duration music recordings. In this work, we propose MusicRWKV, a multimodal music representation learning framework built upon the RWKV architecture. MusicRWKV jointly models waveform, spectrogram, and metadata information within a unified self-supervised learning framework. To improve efficiency for long musical sequences, we introduce an invertible compression module that reduces sequence length while preserving reconstruction capability. A FiLM-based conditioning strategy is employed to facilitate modality-aware representation learning without introducing expensive cross-attention operations. We further investigate the integration of Soft ROSA as a differentiable retrieval mechanism for neural-symbolic interaction. Experimental evaluation across a diverse collection of Music Information Retrieval (MIR) benchmarks demonstrates that the proposed RWKV-based architecture can be effectively adapted to heterogeneous music understanding tasks. A controlled pretraining ablation shows that Soft ROSA improves validation performance, while layer-wise telemetry analysis reveals a progressive transition from distributed retrieval to selective retrieval in deeper network layers. These results suggest that RWKV provides a promising architectural alternative to Transformer-based models for multimodal music understanding.

cs.SD · cs.AI · eess.AS

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.16051

Human-Likeness and Conversational Preference Diverge for Synthesized Agent Speech in Customer-Service Calls

Jeffrey M. Girard, Antony D'Avirro, Jason Z. Zheng, Jacqueline R. Vertino, Benjamin Peloquin

原文摘要

Listeners can judge an agent's speech on how human it sounds (human-likeness) and on how much they would want to talk to it (conversational preference). Listening tests rarely ask about conversational preference, so the relation between the two judgments is unclear. We compared them in paired-choice tests on 120 turns that actors improvised as customer-service representatives. On each trial, two agents spoke the same turn. In Study 1, both agents came from six commercial text-to-speech systems, each with eight voices. In Study 2, one was the actor's original recording and the other a system. Some systems ranked higher on human-likeness, and others on conversational preference. Listeners chose systems over actors in 19% of human-likeness trials but 36% of conversational preference trials (and 49% for the most preferred system). On both judgments, variation among a system's voices was about three-quarters of that between systems. Of three automatic naturalness scores, two predicted lower human-likeness, and none predicted conversational preference. In this setting, listeners tended to favor faster speech and fewer pauses between clauses on both judgments, and more pitch variation for conversational preference. Thus, measuring only human-likeness or testing one voice per system can misjudge which agents listeners would rather talk to.

cs.CL · cs.HC · eess.AS

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2611.06262

SonoDyno: A Sonic Dynamometer for Engine Speed and Load Estimation from Audio

Robin Doerfler, Beat Rossmy

原文摘要

Controllable engine sound synthesis, Active Sound Design and engine order analysis all build on the relationship between the sound of an engine and its momentary operating state: its speed and its load. Recordings of engine sound are abundant in archives, sample packs and online video, yet they rarely carry time-aligned annotations of that state, which limits their applicability to controllable engine sound generation. We present SonoDyno, which estimates the engine speed, the cylinder count and an engine-normalized load intensity from a microphone signal alone, with no tachometer, no bus access and no prior knowledge of the engine configuration. A convolutional scorer of 124,302 parameters reads the energy at the engine orders, the harmonics of half the rotation frequency, in each frame at three time-frequency resolutions to jointly score cylinder count and speed, and a decoder imposes the rotational dynamics of the engine on the speed trajectory. A second head estimates the load intensity, the position of the torque between the engine's drag floor at 0 and its full-load ceiling at 1, from the order levels at the decoded speed. On recordings of engines it has never heard, with microphone position, engine type and cylinder count unknown, SonoDyno places 93 to 97% of the frames within 2% of the true speed, identifies the cylinder count with 80% accuracy and confuses the octave on fewer than 2% of the frames. On an unseen engine of a known type, the load intensity has a mean absolute error of 0.12, and the sign of the torque, driving or braking, is right on 91% of the frames. General pitch trackers fall behind even when the harmonic order of each engine is supplied: on our validation recordings and at a tolerance of 5%, the best of them places 62% of the frames within the tolerance, against 96% for SonoDyno.

cs.SD · eess.AS · eess.SP

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2603.27998

HRIR-Former: Grid-Free Time-Domain Reconstruction of Head-Related Impulse Responses with a Spatially Encoded Transformer

Shaoheng Xu, Chunyi Sun, Jihui Zhang, Amy Bastine, Prasanga N. Samarasinghe, Thushara D. Abhayapala, Hongdong Li

原文摘要

Individualized head-related impulse responses (HRIRs) enable binaural rendering, but dense per-listener measurements are costly. We address HRIR spatial up-sampling from sparse per-listener measurements: given a few measured HRIRs for a listener, predict HRIRs at unmeasured target directions. Prior learning methods often work in the frequency domain, rely on minimum-phase assumptions or separate timing models, and use a fixed direction grid, which can degrade temporal fidelity and spatial continuity. We propose HRIR-Former, a time-domain, grid-free binaural Transformer for reconstructing HRIRs at arbitrary directions from sparse inputs. It uses sinusoidal spatial features, a Conv1D refinement module, and auxiliary interaural time difference (ITD) and interaural level difference (ILD) heads. On SONICOM, it improves normalized mean squared error (NMSE), cosine distance, and ITD/ILD errors over prior methods; ablations validate modules and show minimum-phase preprocessing is unnecessary.

eess.AS · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.15464

RSE-TTS: Compact Text-to-Speech through Shared Text-Aware Reference Encoding and Generation

Ming Ma, Mingjie Chen, Yuanzhong Zheng, Hengyao Bao, Hailin Yu, Shijing Yu, Hao Qiu, Gongfan Cao 等 11 位作者

原文摘要

Can a speech generator also serve as an effective voice encoder? We show that a pretrained autoregressive text-to-speech (TTS) backbone can be repurposed for reference compression, improving speaker similarity in subsequent synthesis. Jointly encoding reference speech and its transcript further strengthens this capability. Controlled ablations and activation interventions trace the text benefit through intermediate acoustic representations into the compressed vectors that condition generation. Building on these findings, we introduce RSE-TTS, whose Reconstructive Speaker Encoder (RSE) shares the generation backbone and compresses paired reference text and speech into eight continuous vectors. Sharing reduces encoder-generator parameters by 48% relative to independent pretrained encoding. Large-scale English training yields a 118M-parameter generator that improves both word error rate and speaker similarity over the corresponding official MOSS-TTS-Nano cloning modes on all three English benchmarks. Combined with acoustic continuation, RSE-TTS approaches the speaker similarity of substantially larger systems while achieving lower WER than the evaluated CosyVoice 3 and MOSS-TTS-Local models. Audio samples are available at https://voxnexusai.github.io/RSE-TTS/.

cs.SD · eess.AS

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2605.06407

WavCube: Unifying Speech Representation for Understanding and Generation via Semantic-Acoustic Joint Modeling

Guanrou Yang, Tian Tan, Qian Chen, Zhikang Niu, Yakun Song, Ziyang Ma, Yushen Chen, Zeyu Xie 等 15 位作者

原文摘要

Integrating speech understanding and generation is a pivotal step toward building unified speech models. However, the different representations required for these two tasks currently pose significant compatibility challenges. Typically, semantics-oriented features are learned from self-supervised learning (SSL), and acoustic-oriented features from reconstruction. Such fragmented representations hinder the realization of truly unified speech systems. We present WavCube, a compact continuous latent derived from an SSL speech encoder that simultaneously supports speech understanding, reconstruction, and generation. WavCube employs a two-stage training scheme. Stage 1 trains a semantic bottleneck to filter off-manifold redundancy that makes raw SSL features intractable for diffusion. Stage 2 injects fine-grained acoustic details via end-to-end reconstruction, while a semantic anchoring loss ensures the representation remains grounded within its original semantic manifold. Comprehensive experiments show that WavCube closely approaches WavLM performance on SUPERB despite an 8x dimensional compression, attains reconstruction quality on par with existing acoustic representations, delivers state-of-the-art zero-shot TTS performance with markedly faster training convergence, and excels in speech enhancement, separation, and voice conversion tasks on the SUPERB-SG benchmark. Systematic ablations reveal that WavCube's two-stage recipe resolves two intrinsic flaws of SSL features for generative modeling, paving the way for future unified speech systems. Codes and checkpoints are available at https://github.com/yanghaha0908/WavCube.

eess.AS · cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2611.13243

GigaEdge Spench: An SDK for Practical Streaming Speech Enhancement

Artem Sokolov, Artem Khaidarov, Yana Lysova, Alexander Gladkikh, Vitaliy Yulis

原文摘要

We present GigaEdge Spench, an open-source SDK for streaming speech enhancement on CPU servers, mobile devices, and in browsers. Applications can build enhancement channels from neural denoising, normalization and dynamics blocks, sample-rate conversion, and acoustic echo cancellation where supported. These stages can be combined, while normalization can also run without a neural model. The release includes an 8 kHz INT8 denoiser and discriminative and hybrid 48 kHz FP32 models. A shared C++ core supports ONNX Runtime and TensorFlow Lite through native interfaces, Android and iOS SDKs, and WebAssembly integration. Independent stream contexts share model resources, and consecutive frames from one stream can be processed in a single inference call. We evaluate the released models on EARS-WHAM v2 and compare their quality and processing costs with RNNoise and DPDFNet2. All tested native and Web configurations process files faster than real time, although larger chunks require more time to collect input. We also measure how many streams a shared model can process on a CPU. Our contribution is a configurable deployment system and an experimental study of how model choice, chunk size, and execution environment affect its use. All reported SDK timing uses ONNX Runtime. Index Terms: speech enhancement, streaming inference, cross-platform deployment, temporal frame grouping, computational efficiency

cs.LG · eess.AS

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.14929

Speech Tokenizers are Vulnerable: Transferable Semantic Attack and Robust Tokenizer

Zhisheng Zhang, Yifan Mi, Yixuan Zhou, Haiyun Li, Shengbo Cai, Dongfu Song, Ziang Xu, Jie Hao 等 9 位作者

原文摘要

Speech tokenizers serve as speech front ends, encoding input speech signals into representations for downstream models, such as automatic speech recognition (ASR) systems and large audio-language models (LALMs). In this paper, we find that speech tokenizers are fragile at the semantic level: small perturbations can disrupt their semantic encoding process and consequently cause severe degradation in downstream ASR and LALM performance. To study this vulnerability, we propose T-SemAttack, a transferable semantic attack based on a semantic encoder ensemble. By jointly drifting the representation spaces of multiple semantic encoders, T-SemAttack effectively disrupts the semantic content of speech while preserving perceptual quality, thereby inducing error features. We further analyze layer-wise representation drift and cross-model gradient alignment. Our findings show that perturbations can induce substantial shifts in deep semantic representations and that the input gradients of different surrogate encoders exhibit directional alignment. These observations suggest that shared adversarially sensitive directions in the input space may be a key factor underlying cross-model attack transferability. Building on these, we introduce ROSETok, a robust speech tokenizer that combines noisy training with robust semantic distillation to improve reconstruction fidelity and downstream SpeechLLM robustness. Extensive experiments on multiple datasets, 9 open-source and black-box ASR systems, and 4 LALMs demonstrate that T-SemAttack achieves strong transferable attack performance, while the proposed Robust Speech Tokenizer exhibits robustness under high-fidelity reconstruction and downstream tasks. Our code and demo are available at https://wxzyd123.github.io/t-semattack.

eess.AS · cs.AI · cs.CR · cs.SD

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.14908

Hy ASR 3.0 Technical Report

Aoxiong Yin, Auden Li, Chao Huang, Chuanyang Hu, Eitan Wang, Ethan Zou, Fangming Feng, Guangyu Chen 等 63 位作者

原文摘要

We present Hy ASR 3.0, a large language model (LLM)-based automatic speech recognition (ASR) system that delivers strong recognition accuracy, supports flexible contextual conditioning, and maintains robust recognition performance across diverse speech conditions. Hy ASR 3.0 supports Mandarin Chinese, English and 13 Chinese dialect groups, including Cantonese. Among the evaluated systems, it achieves the lowest aggregate error rates on public Mandarin Chinese, English, and Cantonese benchmarks and the lowest error rates on the corresponding internal test sets, alongside first-tier dialect recognition performance. Beyond audio-only transcription, Hy ASR 3.0 supports multiple forms of external context, including hotword lists, question--answer dialogues, and summaries, enabling the use of task-relevant information during recognition. The model also leverages intra-utterance context to disambiguate homophonic and phonetically similar keywords using information from other parts of the same utterance, without requiring a separate contextual input. Evaluations demonstrate strong contextual recognition in both external-context and intra-utterance settings, as well as robust recognition of long-form speech, children's and older adults' speech, low-volume speech, and speech in noisy environments. This report describes the model architecture and training methodology and presents evaluations across recognition accuracy, context utilization, and robustness.

cs.CL · eess.AS

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.14801

Distributionally Robust BEST-RQ For Speech Representation Learning

Muhammad Khan, Xiaodong Cui, Brian Kingsbury, George Saon

原文摘要

BEST-RQ, based on random projection and quantization, has been a well-established self-supervised learning approach to learn speech representations from large-scale unlabeled audio due to its high performance and efficiency. However, in BEST-RQ representations learned by random quantization can be sensitive to the choice of quantization targets. To address this issue, we propose a distributionally robust extension of BEST-RQ that seeks representations robust across multiple quantization configurations. Specifically, we define groups by codebooks and formulate the training as a group Distributionally Robust Optimization (DRO) problem. Rather than optimizing solely for the average loss across groups, our approach adaptively reweights groups to emphasize those performing poorly. Additionally, we regularize DRO with a Kullback-Leibler divergence to prevent excessive distribution deviation. We evaluate our approach on the Multilingual LibriSpeech dataset and show that the distributionally robust BEST-RQ achieves superior performance over the conventional uniform average across multiple codebooks.

cs.CL · eess.AS · math.OC

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.14774

An LLM Is All You Need: Data-Efficient Adaptation to Full-Duplex, Word-Level Spoken Dialogue

Tairan Chen, Richard Yucheng He, Yihang Liu, Baodong Cao, Chen Xu

原文摘要

Full-duplex spoken dialogue, in which a system listens while it speaks and can backchannel, yield the floor, and be interrupted, is usually obtained by training on hundreds of thousands to millions of hours of speech. This is costly, rewrites the language model that supplies the system's knowledge, and still leaves turn-taking either implicit or decided coarsely rather than word by word. We show that a frozen, general-purpose multimodal LLM can instead be adapted into a word-level full-duplex system with little data. Our system is a thinker-talker cascade on an 80 ms frame clock. The thinker keeps the LLM frozen and trains only lightweight adapters and a few new token embeddings (97M parameters); at every frame it emits one token of agent text, with pauses, yields, and onsets represented as tokens of the same vocabulary and kept legal by a finite-state machine. A four-stage curriculum, each stage anchored to the model's own behavior under the previous format, performs the conversion in about 30 GPU-hours. The talker reads only the thinker's discrete tokens, so the two halves are trained independently. On Full-Duplex-Bench and InteractionBench, the system is competitive with open-source full-duplex systems built through large-scale speech training: among them, it is the least likely to take the floor when the user merely pauses, backchannels most often, and leads in spoken-task accuracy, groundedness, and coherence over long conversations.

cs.CL · cs.SD · eess.AS

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.14443

评估自动钢琴转录数据集中的对齐：评分方法的基准测试

Assessing Alignment in Automatic Piano Transcription Datasets: A Benchmark for Scoring Methods

Lucas de Souza Brandão, Lucas Gris, Joaquim Cavalcante, Chukwuemeka Nkama, Andrea Poltronieri

AI 速读 基于摘要

提出一个通过受控MIDI损坏来评测音频-MIDI对齐评分方法的框架，并引入基于AMT模型的STI方法，与现有方法进行基准比较。

cs.SD · eess.AS

arXiv ↗PDF ↗
查看原文摘要

Automatic Music Transcription (AMT) relies on paired audio--MIDI datasets, implicitly assuming near-perfect temporal alignment. However, widely used piano datasets have been shown to contain alignment errors that may bias model evaluation and training. We propose a framework for evaluating audio--MIDI alignment scoring methods through controlled temporal, pitch, and structural MIDI corruptions. We also introduce STI, a scoring method built on AMT models, and benchmark it against existing alignment scoring approaches. Our results show that STI, together with methods based on CENS and spectrograms, achieves the strongest overall performance, with STI exhibiting the highest sensitivity to fine-grained temporal misalignments, the main alignment error in AMT datasets. We then apply the best-performing methods to widely used piano AMT datasets and identify trends in their relative alignment quality. These results support more informed dataset curation, model training, and AMT evaluation.

新论文 · v1arXiv:2611.14421

Earlybird：2900万参数下的预测式流式语音识别

Earlybird: Predictive Streaming Speech Recognition in 29M Parameters

Sahil Mahendrakar

AI 速读 基于摘要

作者提出一个29M参数的流式语音识别器Earlybird，通过随机截断音频训练，使其能在词尚未说完时提前预测整词并给出置信度。

cs.SD · cs.CL · eess.AS

arXiv ↗PDF ↗
查看原文摘要

Streaming speech recognizers output a word only after the speaker has finished it, typically 90-400 ms later. For a voice agent every millisecond counts: the sooner the words are outputted, the sooner the agent can start a search or plan its reply. We present Earlybird, a 29M-parameter streaming recognizer (16.8 MB at 4 bits). Earlybird learns from audio cut off at random points, including partway through a word, and is asked to predict every word that is due by then in full, plus a guess of the next two words. Its raw token probabilities give a confidence for every word including the one still being spoken. On clean audiobook speech (LibriSpeech test-clean), the median word is predicted correctly 90 ms before the speaker finishes it, while released streaming models get it 90 to 400 ms after it ends. Earlybird is the earliest on all six test sets we time, covering audiobooks, TED talks, public speech, earnings calls, and synthetic speech. On the same audiobooks, 69% of words are predicted correctly before they end, against 3-25% for the other models. For words still being spoken, confidence is calibrated, with an expected calibration error (ECE) of 3.7 to 5.5. On the eight English Open ASR Leaderboard sets, it is more accurate than Moonshine Tiny on every set and roughly ties Whistle, a model of the same size.

新论文 · v1arXiv:2611.14003

Spatio-Temporal Calibration of a Vehicle-Mounted Microphone Array for Autonomous Driving

Marvin Seegert, Richard Taupitz, Korbinian Moller, Johannes Betz

原文摘要

Microphones let an autonomous vehicle sense into areas its cameras and LiDAR cannot see, by picking up the sound that other traffic participants emit. Locating those sources and fusing them with the visual perception requires the microphone array to be calibrated and registered relative to the vehicle's other sensors. Existing calibration methods assume small, unobstructed indoor rigs, whereas a vehicle body casts an acoustic shadow that straight-line propagation cannot represent, and to our knowledge no calibration of an array distributed around a vehicle has been reported. We propose a pipeline that uses a balloon burst as a single cross-modal target, which the LiDAR locates in metric coordinates, every microphone hears as a sharp onset and a camera times. Each observation is weighted by its own timing variance, which discounts shadowed arrivals instead of discarding them and needs no model of the vehicle. On a full-size vehicle with eight microphones over a 4.9 m aperture, the held-out range residual of 6.37 cm approaches the 6.07 cm floor set by the onset timing noise, and the calibrated array tracks a moving source to within a few degrees of the vehicle's own LiDAR. A dozen bursts spread around the vehicle suffice, and spreading them matters more than adding more.

cs.SD · eess.AS

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.13990

On the information rate of phones, syllables and words

Herman Kamper

原文摘要

Which linguistic units minimise the bitrate required to encode speech content: phones, syllables, words, maybe larger units, or something in between? This question has implications for the development of speech and language systems and theories of human language processing. Using a transcribed corpus of 960 hours of English audiobooks, I compare the information rates of phones, syllables, words, multi-word combinations and intermediate tokenisations. While all units are equivalent under a perfect language model, it makes sense to compare these under limited modelling budgets; I use $n$-gram and transformer language models. The lowest bitrates are achieved using words, with slightly longer units sometimes performing best. The results show that, when modelling capacity is limited, units larger than syllables but smaller than long multi-word phrases provide the most efficient representation in terms of information rate.

cs.CL · eess.AS

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.29164

XingChen-UniFE: A Unified Generative Speech Front-End via Complementary Acoustic-Semantic Modeling

Kelan Kuang, Qihan Hu, Jian Kang, Jie Li, Yongxiang Li, Yi Liu

原文摘要

This paper presents XingChen-UniFE, a unified large audio language model (LALM) for general speech restoration (GSR) and target speaker extraction (TSE). The proposed model employs complementary semantic and acoustic representations through a dual-stream encoder, where a self-supervised learning model provides hierarchical semantic features and a learnable acoustic branch captures fine-grained structures from complex spectra. A bidirectional cross-attention module further facilitates information exchange between the two representations. The fused representation conditions the LALM for discrete speech token generation with task and speaker information. Experiments demonstrate strong performance on both GSR and TSE and robust generalization to unseen degradation conditions.

eess.AS · cs.SD

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.13689

Walrus Technical Report

Alkis Koudounas, Osamu Take, Hayato Futami, Quentin Jodelet, Emiru Tsunoo

原文摘要

We present Walrus, a unified 8B speech-and-audio foundation model handling speech recognition (including code-switching and multi-talker speech), speech-to-text and speech-to-speech translation, text-to-speech, emotion recognition, audio captioning, and general audio understanding. Our main contribution is not a new architecture but a data-centric training recipe; we use a standard decoder-only backbone initialized from Gemma 4, so that gains can be traced to the data. For speech generation, Walrus predicts low-bitrate discrete speech tokens, and a lightweight token-to-waveform decoder converts them into high-fidelity audio. The recipe is a three-stage data curriculum that moves from broad multilingual data to multi-talker, code-switching, and domain-specific data. At each stage, task-aware and dataset-aware weighting balances tasks with very different token budgets and protects core skills from forgetting. Trained on over 600M samples covering more than 20 tasks and 100 languages, Walrus leads on most multilingual, multi-talker, and code-switching ASR, speech translation, and audio understanding benchmarks, outperforming both open speech language models of comparable size and dedicated task-specific models. On most benchmarks, it even matches or surpasses the $4\times$ larger Qwen3-Omni's performance. Results suggest that data engineering is a primary driver of quality in multi-task speech and audio modeling.

eess.AS · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.13614

Emilia2: A Large-Scale Dataset for Controllable Long-Form Speech Generation

Qinke Ni, Dekun Chen, Junwen Qiu, Huan Liao, Xu Li, Xingwen Han, Wan Lin, Kunyu Feng 等 17 位作者

原文摘要

Generating long-form speech requires models to keep content, speaker identity, turn-taking, and expressive style consistent over time. Yet most open training corpora contain examples lasting only tens of seconds and open long-form benchmarks cover at most two speakers per recording. We introduce Emilia2, a 229K-hour multilingual dataset for controllable long-form speech generation. From in-the-wild data, Emilia2 constructs short utterances, single-speaker long-form speech, and multi-speaker dialogues, preserving transcripts, speaker identities, pauses, and acoustic transitions. Examples longer than five minutes account for 50.0% of single-speaker long-form hours and 66.8% of dialogue hours. To support expressive control, Emilia2 provides natural language instructions for 4.77M examples and nonverbal-vocalization tags for 933K examples. We also introduce Emilia2-testset, 23 hours of bilingual recordings of up to 30 minutes and four speakers, each speaker paired with reference audio. We train autoregressive (AR) and non-autoregressive (NAR) models on Emilia2. Our 0.6B-backbone NAR model achieves the lowest Chinese CER and English WER on TTSD-Eval (5.27%/9.93%) among compared systems, while our AR model is on par with Qwen3-TTS-12Hz-VD on InstructTTSEval (zh/en: 76.2/78.3 vs. 77.1/77.9). These results validate Emilia2 as a comprehensive data foundation for long-form speech generation and minute-scale expressive control. Demo: https://charlesnii.github.io/Emilia2-demo/.

eess.AS · cs.SD

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2610.35581

Integrate the Reward, Don't Sample It: Window Likelihood for Interruptions and Backchannels in Full-Duplex Speech Models

Guoqiang Hu, Lichao Zhang, Yizhou Peng, Yuhao Luo, Haifeng Hu, Zhongqian Sun, Wei Yang, Eng Siong Chng

原文摘要

When a user talks over a full-duplex dialogue model, it must decide whether the user is interrupting (stop) or backchannelling (carry on). Post-training can improve this discrimination or change how readily it stops. In our GRPO runs the second dominates: no configuration lowers both error rates on more than one model, and two released checkpoints discriminate no better than their bases. We trace this partly to how the reward is estimated. The probability of stopping inside a short window is a known function of a few stop probabilities the policy emits, yet GRPO estimates it by sampling. A group of rollouts that all agree gives no gradient, even when all fail; the push on the stop threshold is strongest where the model is unsure, whichever outcome is right; and whole-rollout credit reaches decisions the reward never judges. Confining the credit to the window removes only that leakage. Window likelihood (WL) samples the words but integrates over the stop decisions, so the probability of the correct outcome is computed exactly and its logarithm maximised. On HumDial-FDBench, across four open models, WL gives the best F1 and lowers both error rates by more than their confidence intervals, without talking into silence.

eess.AS · cs.CL · cs.SD

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.13430

Off-target posterior mass: A training-free diagnostic for fine-tuning spoken language diarization probes

Zhuoyan Tao, Aditya Kommineni, Shrikanth Narayanan

原文摘要

Spoken Language Diarization (SLD) models provide frame-level language information of a speech recording. SLD modeling pipeline includes feature extraction from pre-trained speech models followed by clustering. Fine-tuning pre-trained speech models has shown performance gains for SLD with its magnitude being dataset dependent. Off-target Posterior Mass (OTPM), computed based on posterior language identification probabilities is proposed as a diagnostic metric to forecast performance gains from SLD fine-tuning. OTPM obtains positive correlation with probe gains, 0.56 on unseen synthetic bilingual data and 0.23 on real-world corpora. These results indicate promising potential of OTPM as a diagnostic metric for SLD.

eess.AS · cs.CL · cs.SD

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2604.17958

MINT-Bench: A Comprehensive Multilingual Benchmark for Instruction-Following Text-to-Speech

Huakang Chen, Jingbin Hu, Liumeng Xue, Qirui Zhan, Wenhao Li, Guobin Ma, Hanke Xie, Dake Guo 等 15 位作者

原文摘要

Instruction-following text-to-speech (TTS) has emerged as an important capability for controllable and expressive speech generation, yet its evaluation remains underdeveloped due to limited benchmark coverage, weak diagnostic granularity, and insufficient multilingual support. We present \textbf{MINT-Bench}, a comprehensive multilingual benchmark for instruction-following TTS. MINT-Bench is built upon a hierarchical multi-axis taxonomy, a scalable multi-stage data construction pipeline, and a hierarchical hybrid evaluation protocol that jointly assesses content consistency, instruction following, and perceptual quality. Experiments across ten languages show that current systems remain far from solved: frontier commercial systems lead overall, while leading open-source models become highly competitive and can even outperform commercial counterparts in localized settings such as Chinese. The benchmark further reveals that harder compositional and paralinguistic controls remain major bottlenecks for current systems. We release MINT-Bench together with the data construction and evaluation toolkit to support future research on controllable, multilingual, and diagnostically grounded TTS evaluation. The leaderboard and demo are available at https://aslp-lab.github.io/MINT-Bench-Demo/

eess.AS · cs.SD

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12527

WaveGen: Truly End-to-End Waveform Generation via Internal Spectral Trajectory Learning

Sang-Hoon Lee, Ha-Yeong Choi

原文摘要

This work proposes WaveGen, a fully end-to-end diffusion-based text-to-waveform framework based on Spectral Forcing, which jointly learns an internal spectral trajectory and the target waveform trajectory within a single model, enabling direct end-to-end text-to-waveform generation. Unlike recent TTS pipelines, WaveGen does not rely on separately trained or externally decoded intermediate representations, such as neural audio codecs, Mel-spectrogram vocoders, or VAE-based latent models. In addition, WaveGen performs internal text-speech alignment within a single model, eliminating the need for external alignment modules such as duration predictors. To preserve a truly end-to-end formulation, our framework further avoids dependence on self-supervised representations, such as BERT-based text embeddings and Wav2Vec 2.0-based semantic representations, as well as semantic distillation methods for training optimization. To achieve this, we carefully design Spectral Forcing, a model architecture with disentangled diffusion heads, and a training objective based on a multi-scale spectral $v$-loss. We further validate the scalability of end-to-end TTS models by scaling the model size from 0.1B to 0.6B parameters. Using a single architecture, open-source data, and one-stage training, without external modules such as duration predictor or REPA, WaveGen demonstrates fully end-to-end waveform generation achieving state-of-the-art performance on the Emilia training dataset. In particular, WaveGen achieves a WER of 1.47 and a SIM of 0.75 on the Seed-en benchmark. Source code and checkpoints are available at https://github.com/sh-lee-prml/WaveGen

cs.SD · cs.AI · eess.AS · eess.SP

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12406

REAL: Reward Modeling through Evidence-Linked Analysis of Lifelikeness in Generated Speech

Yaodong Song, Jie Lian, Xusheng Yang, Genliang Zhao, Hongjie Chen, Jian Kang, Zhen-Hua Ling, Jie Li 等 9 位作者

原文摘要

As modern text-to-speech systems become increasingly intelligible and natural, evaluation is shifting from detecting gross synthesis failures to identifying subtle departures from lifelike human delivery. Because large-scale listening tests are costly and slow, speech reward models can serve as scalable automatic judges for comparing generated utterances, curating data, and supporting iterative TTS development. Existing speech reward models can predict scores or pairwise preferences and may generate natural-language rationales, but their supervision is typically defined at the utterance or pair level. The connection between local acoustic phenomena and the final preference therefore remains implicit. We introduce REAL, a framework for Reward modeling through Evidence-linked Analysis of Lifelikeness in generated speech. Its central idea is to supervise not only the final pairwise preference but also a structured account of its supporting observations, linking local audio spans to perceptual phenomena, affected dimensions, and the overall judgment. Specifically, REAL-Jury constructs these evidence-linked reports through timestamp alignment, multi-model analysis, and dispute-oriented deliberation. REAL-Judge learns to generate them from paired audio inputs. REAL-Eval assesses preference prediction in two tiers, from salient perceptual differences to subtle distinctions among strong TTS systems. Instantiated with MiMo-Audio, REAL-Judge improves pairwise preference accuracy over its backbone by 16.01 and 11.41 percentage points on the two tiers, respectively, and outperforms the open-source speech reward models SpeechJudge and UniSRM on both tiers.

eess.AS · cs.SD

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12375

AudioCodex: Streaming Autoencoders for Live Audio

Marco Pasini, Javier Nistal, Cyran Aouameur, Stefan Lattner, George Fazekas

原文摘要

We introduce AudioCodex, a fully causal autoencoder for live generative audio applications that compresses stereo audio into compact continuous latent sequences at ~10 Hz. It combines a Mean Flow magnitude autoencoder with latent hypersphere projection, bounded-noise augmentation, representation alignment, and adversarial phase recovery to learn representations that support both efficient downstream generation and high-quality reconstruction. Through extensive ablations, we show that AudioCodex outperforms the state-of-the-art in terms of generation-reconstruction trade-off, whilst remaining strictly causal in both encoding and decoding. This work paves the way towards truly real-time interactive and streaming generative audio applications.

cs.SD · cs.LG · eess.AS

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12124

SCLAP: Temporal structural alignment improves localization in long-form audio

Yacouba Kaloga, Ina Kodrasi

原文摘要

Audio-text models such as Contrastive Language-Audio Pretraining (CLAP) typically pool frame and token sequences uniformly into global embeddings. This discards temporal structure and impairs both retrieval and localization, especially for short events in long recordings. We introduce SCLAP, a sequence-level contrastive framework that aligns audio frames and text tokens using monotone one-dimensional optimal transport. Its learned marginals and transport plan provide fine-grained temporal supervision from clip-level captions alone. On short-form audio (i.e., Clotho and TACOS datasets with 15-30s recordings), SCLAP remains competitive for retrieval while improving zero-shot localization over CLAP and Text-aware Attention Pooling. On long-form audio (i.e., CASTELLA dataset with up to 300s recordings), SCLAP improves retrieval performance and its localization advantage grows as the event-to-context ratio decreases. At inference, SCLAP requires no optimal transport computation since retrieval can use pooled embeddings and localization can use inexpensive frame-text cosine scores.

eess.AS · cs.SD

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12119

Discriminator-free adversarial domain adaptation for robust speech classification

Yacouba Kaloga, Ina Kodrasi

原文摘要

The performance of speech classification systems such as speech emotion recognition or pathological speech detection can degrade substantially when there is a domain shift between training and test data. This shift can arise due to e.g., differences in recording setup, acoustic environment, speaker, language, or speaking style. Adversarial domain adaptation addresses this problem by encouraging representations that are less sensitive to domain-specific differences. However, many adversarial approaches rely on training a separate domain discriminator, which can be unstable and difficult to train. In this paper, we propose to use a discriminator-free adversarial learning network (DALN) for unsupervised domain adaptation in speech classification tasks. DALN reuses the task classifier as an implicit discriminator and aligns domains through the nuclear-norm Wasserstein discrepancy on classifier outputs. We extensively evaluate the performance of DALN on intra-lingual, cross-lingual, and noisy adaptations for speech emotion recognition and pathological speech detection. Extensive experimental results show that DALN generally achieves the best performance across the considered tasks and domain shifts, outperforming discriminator-based domain adaptation approaches.

eess.AS · cs.SD

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12061

BARS: Beat-Adaptive Rap Synthesis via Flow Matching

Taemin Cho, Hector Martel, Enyan Koh

原文摘要

We present BARS (Beat-Adaptive Rap Synthesis), a non-autoregressive rapping voice generation model based on Flow Matching, conditioned on a beat track (instrumental backing), lyrics, and a short reference voice for zero-shot timbre control. Unlike prior autoregressive rap synthesis, where rap pace and output duration are strictly determined by the beat, BARS enables independent control over the rap tempo, allowing flexible-length generation without explicit duration prediction or phoneme-level alignment. We encode lyrics using a pretrained ByT5 encoder with a lightweight Transformer adapter trained jointly with the model, without relying on speech-pretrained text encoders, enabling natural handling of non-standard orthography (e.g., slang, abbreviations) and numerals common in rap lyrics. To address alignment instabilities, such as token skipping or repetition, under strong beat conditioning, we propose Asymmetric Scaled RoPE for cross-attention, which introduces a positional structure to facilitate sequential lyric-audio alignment learning. Furthermore, the model supports partial re-generation via span masking, enabling lyric-level editing while preserving the surrounding vocal style. Experiments demonstrate that the proposed model achieves high lyric fidelity, natural rap flow that is rhythmically coherent with the beat, and robust tempo control across a wide range of rapping styles and speeds.

cs.SD · cs.LG · cs.MM · eess.AS

查看论文
arXiv ↗PDF ↗
版本更新 · v7arXiv:2602.14612

Event-Grounded Question Answering over Long Audio via Structured Retrieval

Kartik Hegde, Arvind Krishna Sridhar, Naveen Vakada, Yinyi Guo, Erik Visser

原文摘要

Answering natural-language questions over multi-hour audio requires reliable event recognition, temporal grounding, and efficient retrieval. We present LA-RAG (Long Audio Retrieval-Augmented Generation), a structured framework that converts audio into timestamped event records, stores them in an event database, and answers questions using intent-aware retrieval and LLM-based generation. LA-RAG supports two deployment settings: offline grounding mode, in which long recordings are pre-indexed for low-latency querying, and inference-time grounding mode, which performs query-conditioned grounding over shorter, open-ended clips. We evaluate LA-RAG on two synthetic 24-hour benchmarks, Home-IoT and Industrial-IoT, and on CASTELLA-QA and SpotSound-Bench-QA, which are QA versions of two public audio temporal grounding benchmarks that we construct. The results demonstrate effective long-audio question answering with low query-time latency after grounding and indexing. They also reveal a substantial gap between event detection and temporal localization in current large audio-language models. Explicit timestamped grounding consistently improves temporal localization while maintaining competitive detection accuracy. These findings establish structured grounding and retrieval as a practical complement to generative audio-language models for deployment-oriented long-audio understanding.

eess.AS · cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗