arXiv_SD
-
DiffSynth-Music: Audio-Conditioned KV-Cache Adapters for Controllable Music Generation
2026-09-11 12:28:28Zhongjie Duan, Shengchuan Gao, Hong Zhang, Yingda ChenAbstract
Text and lyrics specify broad musical characteristics and sung content but offer limited control over musical timing, melody, and reference-based style. We introduce DiffSynth-Music (this https URL), a framework that adds composable audio conditioning to a music synthesis backbone through layer-wise key-value injection. The three template models, Control, Prosody, and Reference, are initialized from the backbone diffusion transformer and trained with conditional flow matching. They support five control types: beats, vocals, accompaniment, prosody, and reference audio. A shared variational autoencoder maps conditioning waveforms into a common latent space, enabling their attention memories to be combined. With the template timestep fixed at the clean-data endpoint and other inputs held constant, each control cache is computed once and reused throughout sampling. Training pairs are derived from music recordings using beat extraction, source separation, vocal resynthesis, and reference-excerpt selection. Single-control evaluations on Mandarin and English songs demonstrate improved adherence across all five control types and better lyric fidelity under vocal conditioning relative to the backbone. Automatic music-quality and instruction-following scores remain broadly comparable to those of the evaluated base models, with metric-specific trade-offs. We release the three template models to support research and creative applications in controllable music generation.
Abstract (translated)
URL
https://arxiv.org/abs/2609.12774
PDF
-
StepAudio 3 Music Technical Report
2026-09-11 10:02:11Chengli Feng, Zhiyue Wu, Jiahao Song, Zheqi Dai, Boyang Wang, Ruibin Yuan, Junming Gong, Wenxiao Zhao, Jing Guo, Gang Yu, Xiangyu Zhang, Xuerui Yang, Chao YanAbstract
We introduce StepAudio 3 Music, a large-scale, long-form music generation model that supports explicit musical planning and open-domain text-controlled generation. The StepAudio Music Tokenizer represents audio as a 50-Hz stream from a 65536-entry single codebook, using semantically informed self-supervised and multi-task training to preserve musical structure and reconstruction-relevant information. A flow-matching diffusion Transformer (DiT) predicts continuous StepAudio VAE latents, which our VAE decoder converts into 48-kHz audio. This discrete-continuous design is guided by comparisons of single-codebook VQ, Semantic and Acoustic RVQ, and different DiT configurations. For explicit planning, a Mixture-of-Experts autoregressive model uses ABC notation to produce an intermediate arrangement plan (ABC-CoT) before predicting music tokens, making harmony, rhythm, and melodic structure part of the generation context. A progressive training curriculum and supervised fine-tuning support song and instrumental generation, accompaniment generation from dry vocals, and cover-song synthesis for up to 5 minutes and 30 seconds. With reinforcement learning via direct preference optimization (DPO), the final model achieves the highest AudioBox Content Enjoyment, Content Usefulness, and Production Quality scores and the highest MuQ-MuLan similarity among the evaluated systems, with competitive SongBench results. On the preliminary Artificial Analysis Music Arena Vocals leaderboard, it obtains a Quality Elo of 1105, behind only Suno V5.5 and Mureka and ahead of Suno V5, MiniMax models, and other systems. Audio demonstrations are available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2609.16034
PDF
-
Direct Preference Density Alignment for Conversational Audio Equalization
2026-09-11 09:05:08Ioannis Stylianou, Sven Ewan Shepstone, Jon Francombe, Pablo Martinez Nuevo, Zheng-Hua TanAbstract
Large Language Model alignment typically relies on learned proxy reward models, which significantly increase the memory footprint during training and are notoriously prone to instability and reward hacking. While offline methods like Direct Preference Optimization (DPO) bypass the reward model, they lose the ability to perform online exploration. If no optimization constraints are applied, this can lead to format collapse in bounded, continuous spaces. To resolve this, we propose Direct Preference Density Alignment: An alternative framework that removes the need for a learned proxy reward model while strictly preserving the benefits of online reinforcement learning. We leverage large-scale user data (approximately 90,000 samples) to construct non-parametric preference density maps, establishing an empirical reward surface. In addition to removing the reward model, Direct Preference Density Alignment enables the combination of the online structural grounding of Group Relative Policy Optimization (GRPO) with the targeted offline refinement of DPO. We show that this GRPO+DPO combination achieves the highest performance, and in a blind audio equalization listening test, enables a 1.5B-parameter model to achieve perceptual parity with a carefully prompt-engineered GPT-4o mini baseline, using only a fraction of the inference compute.
Abstract (translated)
URL
https://arxiv.org/abs/2609.12607
PDF
-
TokenMapper: A Step Toward Interoperable Speech Token Translation
2026-09-11 08:04:09Tal Kozakov, Tal Rosenwein, Eliya NachmaniAbstract
Neural audio codecs discretize speech into token sequences, but the resulting token spaces differ in vocabulary and codebook structure, preventing direct communication across models. This limitation affects applications such as conversational voice agents and speech to speech translation systems where multiple speech models must interact. As a result, transferring information between speech systems typically requires decoding to waveform audio and re-encoding with a second tokenizer, increasing latency and introducing potential information loss. To address these limitations, we present TokenMapper, a direction aware framework for direct token to token translation between heterogeneous speech tokenizers in the discrete domain. TokenMapper supports structurally mismatched token spaces, including mappings between single codebook and multi codebook representations, under a shared effective token rate. Experiments on GLM-4-Voice, MiMi and DualCodec show consistent cross model performance. Specifically, translation WER approaches native reconstructions within 2.5-6.8% absolute WER, human MOS for TokenMapper outputs ranges from 2.29 to 4.39, following the same direction level trends as UTMOS and end to end latency is reduced by 4.8-94.5% relative to waveform bridging, reaching up to 972 ms per utterance. These results provide a practical step toward cross model speech token interoperability without intermediate waveform reconstruction.
Abstract (translated)
URL
https://arxiv.org/abs/2609.12563
PDF
-
Overview and Meta-Analysis of DCASE 2026 Challenge Task 6: Audio Moment Retrieval from Long Audio
2026-09-11 06:30:41Hokuto Munakata, Tatsuya Komatsu, Keisuke Imoto, Taichi Nishimura, Huang Xie, Tuomas VirtanenAbstract
This paper presents an overview of the Detection and Classification of Acoustic Scenes and Events (DCASE) 2026 Challenge Task 6, Audio Moment Retrieval (AMR) from Long Audio. Given a several-minute-long audio recording and a free-form text query, AMR aims to retrieve temporal moments in the recording that match the query, where each moment is represented by a pair of start and end timestamps. This task requires effective cross-modal alignment and long-range temporal modeling. We describe the task definition, the evaluation metrics, the development and evaluation datasets, and a baseline system that combines a pre-trained MS-CLAP feature extractor with a Detection Transformer (DETR)-based moment-detection network. On the development data, the baseline trained on a manually annotated dataset and a synthetic dataset achieved Recall1@0.7 of 13.56%, indicating that AMR in long audio remains a challenging problem. The challenge attracted 21 teams, which submitted 59 systems in total. The three best systems achieved Recall1@0.7 of 48.59%, roughly 3.5 times the baseline score. The results show that strengthening the audio-text feature extractor and the moment-detection network led to substantial performance improvements. Furthermore, the top three teams boosted performance by applying confidence score calibration or ensembling across different temporal resolutions of features.
Abstract (translated)
URL
https://arxiv.org/abs/2609.12484
PDF
-
DriftSE: Speech Enhancement with Generative Drifting
2026-09-10 22:17:31Liang Xu, Diego Caviedes-Nozal, W. Bastiaan Kleijn, Longfei Felix Yan, Rasmus Kongsgaard OlssonAbstract
We propose DriftSE, a novel one-step generative framework for speech enhancement formulated as a latent distribution equilibrium problem. During training, the drifting field aligns the generator's pushforward distribution with the clean speech manifold through drifting in a latent domain. During inference, the drifting process is discarded, enabling one-step generation. We establish that its enhancement quality depends fundamentally on the choice of latent representation. Semantic latents preserve phonetic structure but fail to capture physical acoustic cues, whereas acoustic latents reconstruct the physical signal but risk linguistic hallucination. Therefore, we introduce dual-latent drifting, performing parallel drifting in both semantic and acoustic latents to simultaneously preserve phonetic intelligibility and acoustic fidelity. Additionally, we demonstrate that DriftSE enables fully unpaired training by aligning latent distributions rather than exact point-wise targets. Consequently, DriftSE facilitates cross-dataset learning in the absence of paired noisy-clean samples. Moreover, DriftSE exhibits broad architectural flexibility across different generator backbones. Extensive evaluations on additive denoising and convolutive dereverberation demonstrate robust one-step enhancement across both offline and real-time causal settings. Notably, DriftSE achieves state-of-the-art word error rates across all four evaluated datasets while strictly operating at 1 NFE. Code and audio examples are available online.
Abstract (translated)
URL
https://arxiv.org/abs/2609.12252
PDF
-
Real-Time Music Source Separation on a Low-Power Audio DSP
Abstract
Real-time music source separation is validated on desktop CPUs and GPUs. Does any published system fit the embedded audio hardware it targets? On a commercial audio DSP (2 MB SRAM, 2.07 GMAC/s measured), none does, and the constraints eliminate different models: memory rules out the 16-51 M parameter TasNet/X-UMX family, per-frame compute rules out RT-STT, needing 5.5x the available MAC rate. Parameter count predicts neither: weight reuse spans 1x to 345x. We then build one that fits. Training on continuous rather than block-padded convolution context proves essential: a model scoring 3.93 dB block-wise otherwise collapses to silence within 2 s frame-by-frame. A gated complex FIR deep filter adds a latency knob, gaining 0.38 dB even when strictly causal. It reaches 4.70 dB cSDR on MUSDB18-HQ and runs in 10.43 ms of an 11.6 ms hop, 0.5-0.7 dB behind systems that do not fit.
Abstract (translated)
URL
https://arxiv.org/abs/2609.12201
PDF
-
Neural Multichannel Distant Speaker Diarization with Heavy-tailed Source Separation Model
2026-09-10 19:41:03Sicheng Mao, Baihan Li, Mathieu Fontaine, Anthony Larcher, Roland BadeauAbstract
Distant speaker diarization remains challenging due to difficult acoustic environments, varying numbers of speakers and overlapping speech. Model-driven methods are proposed to exploit the speech source features in multi-channel recordings that help diarization. This paper generalizes a neural model that jointly learns to perform blind source separation and diarization over speech mixtures (neural FCASA) with heavy-tailed models. The popular Gaussian distribution has been applied for variance modeling in the original source separation model, which we replace with two families of heavy-tailed models (Leptokurtic Generalized Gaussian distribution and Student's t distribution) to better capture the heavy-tailedness in speech signals. Thanks to the Gaussian scale mixture model, we are able to unify the proposed method and the original one under the same form of learning objective. Our experiments show consistent large improvements in Diarization Error Rate (DER) and Jaccard Error Rate (JER) compared to the baseline on various corpora.
Abstract (translated)
URL
https://arxiv.org/abs/2609.12154
PDF
-
How Robust Are Neural Audio Codecs for African Speech? A Multi-Task Benchmark and the Limits of Perceptual Quality
2026-09-10 17:40:32Chibuzor Okocha, Christan Earl GrantAbstract
Neural audio codecs enable low-bitrate speech compression and tokenization, yet their robustness on accented and multilingual speech remains under-evaluated. We benchmark seven open-source neural codecs (DAC, EnCodec, FocalCodec, LanguageCodec, SemantiCodec, UniCodec, WavTokenizer) on three African speech datasets afrinames, afrispeech dialog, afrispeech multilingual, reporting signal-level quality (NISQA, UTMOS, ViSQOL, STOI, F0-RMSE) alongside two downstream tasks: automatic speech recognition (ASR) and speaker verification (ASV). Our analysis yields four findings. First, signal metrics differ sharply in downstream validity: reference-based structural/intelligibility measures (ViSQOL, STOI) and prosodic error (F0-RMSE) track ASR and ASV degradation far more reliably than neural mean-opinion-score predictors (NISQA, UTMOS). Second, intelligibility and speaker-identity preservation diverge sharply across architectures, and the apparent identity ranking itself depends on the ASV backend. Third, degradation is strongly domain-dependent and largest for conversational dialog. Fourth, the resulting degradation is partly recoverable: parameter-efficient \emph{codec} adaptation (LoRA, $\sim$1--3\% of parameters) on roughly 35 hours of African speech reduces the compression-induced word-error-rate gap, recovering recognition toward the uncompressed baseline (developed in companion work). These results motivate task-aware, domain-representative, and adaptation-aware evaluation of speech codecs as a prerequisite for inclusive deployment.
Abstract (translated)
URL
https://arxiv.org/abs/2610.00154
PDF
-
Continuous-Time Acoustic Modelling with Neural Controlled Differential Equations
2026-09-10 15:41:32Mattias Cross, Minghui Zhao, Anton RagniAbstract
Text-to-speech (TTS) models commonly address text--speech alignment by expanding phone-level encoder states to frame-level decoder inputs using predicted durations. While this length-regulation step resolves alignment structurally, this use of duration typically changes only where and how often latent states appear, not the values of the states themselves. This paper proposes a continuous-time mechanism for duration-aware acoustic modelling in TTS using neural controlled differential equations (CDEs). We formulate the phone representation as a temporally parameterised control path and use a neural acoustic vector field to produce a continuous-time hidden state whose values evolve with phonetic content and duration-derived timing. The resulting trajectory can be sampled at discrete points and integrated into a standard acoustic decoder pipeline. Objective results contrast CDEs and typical recurrent models. Subjective results suggest that CDE-based models evaluating one phone per step can improve rank-order agreement between synthesised and reference emotion intensity while maintaining comparable emotion-expression quality to a strong baseline. Additional experiments with half-phone step-sizes suggest that temporal resolution changes the trade-off between style tracking and absolute calibration. These results position CDEs as a promising design space for continuous-time and duration-aware style-sensitive TTS.
Abstract (translated)
URL
https://arxiv.org/abs/2609.11725
PDF
-
ZipCodec: Ultra-Low-Frame-Rate Streaming Speech Coding
2026-09-10 14:49:54Luca Della Libera, Cem Subakan, Mirco RavanelliAbstract
Neural audio codecs are a fundamental component of modern speech generation systems. While recent codecs achieve increasingly low bitrates, reducing frame rate remains challenging, as each token must preserve more information while maintaining reconstruction quality. We present ZipCodec, a streaming neural speech codec operating at 6.25 Hz and 0.80 kbps with a theoretical latency of 160 ms. Our approach combines large-scale WavLM distillation with a redesigned transformer-based architecture, a scalar spherical quantizer, and a latency-aware streaming decoder. Experiments show that ZipCodec substantially outperforms existing streaming codecs at comparable bitrates in both reconstruction and downstream tasks, while operating at a significantly lower frame rate. Despite its 842M parameters, ZipCodec achieves real-time single-stream inference on a consumer-grade CPU. Demo samples, code and checkpoints are available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2609.11642
PDF
-
Post-Training Zero-Shot TTS for Fine-Grained Emotion and Duration Control via Natural Language
2026-09-10 13:24:06Lianru Gao, Yujie Guo, Yong QinAbstract
Audiobook narration, conversational agents, and audiovisual dubbing require speech that conveys changing emotions and adapts its pacing within a single utterance. But most existing TTS systems typically rely on utterance-level style conditioning, making such fine-grained control difficult to achieve. In light of this, and inspired by the success of post-training in large language models, we propose a unified post-training framework that equips pretrained text-to-speech models with natural-language control over segment-level emotion and duration. Supervised fine-tuning establishes instruction-conditioned speech generation, while reinforcement learning with group relative policy optimization refines control accuracy using emotion and duration rewards alongside content and speaker preservation objectives. By reusing the pretrained architecture, our approach avoids additional inference-time control modules. Experiments demonstrate significantly improved fine-grained controllability while maintaining speech intelligibility and speaker identity, highlighting post-training as a practical approach to extending existing speech synthesis models.
Abstract (translated)
URL
https://arxiv.org/abs/2609.11523
PDF
-
Investigating catastrophic forgetting in sound event classification
2026-09-10 12:16:56Riccardo Casciotti, Annamaria MesarosAbstract
This work investigates a number of approaches to prevent catastrophic forgetting in class incremental learning scenarios for sound event classification tasks. We analyze the problem using architectural and regularization approaches, using FSD50K and AudioSet datasets. We design incremental stages and solutions that selectively protect the kernels of the network from weight updates to prevent catastrophic forgetting, and a dynamic head solution that expands itself each time a new task is learned. The findings show that catastrophic forgetting mainly happens in deeper layers, in particular in the classifier head. For the studied in-domain sound classification problem, the solution that seems to alleviate catastrophic forgetting and is the most efficient is a full freezing of the feature extractor with a fine-tuning of the dynamic head classifier, showing little to no forgetting and great training stability, and a good balance between memory-stability and learning plasticity.
Abstract (translated)
URL
https://arxiv.org/abs/2609.11447
PDF
-
X-AuT: Progressive Audio-Encoder Compression for Speech LLMs with Cross-Scale Distillation
2026-09-10 11:43:17Haojun Zhang, Yi Zou, Min Chen, Qize Yu, Lianrui Fan, Xini Ding, Hao Li, Shuchang Zhou, Xianming Liu, Shiyu HuangAbstract
Reducing audio-encoder depth lowers the inference cost of speech large language models, but removing complete blocks perturbs the embeddings consumed by the decoder and can cause deletion and premature end-of-sequence errors. We introduce X-AuT, a progressive framework that selects layer combinations through short behavioral probes and restores the pruned model through representation alignment, cross-scale distillation, scheduled student-policy supervision, and LoRA finetuning. The language-model backbone remains frozen, while attention LoRA adapters and the tied output embedding adapt during distillation. Training uses the highest-agreement tier from a transcript-consistency pipeline, followed by source reweighting during finetuning. On ten public Chinese--English benchmarks, compressing Qwen3-ASR-0.6B from 18 to 16 audio-encoder layers reduces macro-average error from 5.61% to 5.27%. The 14-layer model reaches 5.75% with 20.7% fewer audio-tower parameters. Under the matched recipe, the 1.7B teacher yields 5.55% mean error, compared with 8.45% for self-distillation, and progressive 18$\rightarrow$14 pruning outperforms direct pruning (5.75% vs. 6.73%). These single-run results establish two practical operating points and show that the accuracy effects vary across benchmarks. Project website: this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2609.11412
PDF
-
Copying Versus Randomization in Lempel-Ziv Music Synthesis
Abstract
We utilize Lempel-Ziv universal compression for music note generation. We control the algorithm's tendency to over-copy or under-copy training data by manipulating the average sequence length saved in the dictionary.
Abstract (translated)
URL
https://arxiv.org/abs/2609.11353
PDF
-
EConv-TasNet: Efficient Conv-TasNet for Effective Speech Separation
2026-09-10 10:20:47Pei-Chun Chang, Chuan-Yi LiuAbstract
Conv-TasNet has served as a strong baseline for time-domain speech separation, and many studies have extended it with advanced architectures such as dual-path networks, U-Nets, and attention mechanisms. However, these methods often introduce high computational cost and complexity, limiting their deployment in resource-constrained scenarios. To address this issue, we propose eConv-TasNet, an efficient variant of Conv-TasNet that improves both effectiveness and efficiency without relying on resource-intensive modules. The proposed model consists of a group-wise early-splitting (GES) module and a multi-group feature aggregation (MGFA) module. GES generates discriminative speaker embeddings at intermediate stages, while MGFA progressively aggregates these group-level representations for refined mask estimation. Experimental results show that eConv-TasNet reduces model size by 22.4%, accelerates inference by 18.9%, and improves SI-SNRi by 14.0%-28.0% across three public benchmarks. Moreover, it achieves competitive performance compared with state-of-the-art methods while requiring significantly fewer parameters and lower inference cost. These results demonstrate a favorable efficiency-effectiveness trade-off for edge deployment.
Abstract (translated)
URL
https://arxiv.org/abs/2609.11342
PDF
-
Preference Optimization with LALM Feedback for Continuous Autoregressive Non-Verbal Vocalization Generation
2026-09-10 08:54:28Jingbin Hu, Qirui Zhan, Yuang Cao, Ziyu Zhang, Yunxiang Chen, Houdun Liu, Shuo Feng, Bengu Wu, Lei Xie, Liumeng XueAbstract
We propose a preference optimization framework with Large Audio-Language Model (LALM) feedback for controllable non-verbal vocalization (NVV) generation in continuous autoregressive speech models. To construct preference data without human preference annotation, we build a bilingual prompt corpus by combining NVV-injected real transcripts with LLM-generated semantically aligned prompts, perform stochastic model rollouts, and use a LALM to rank candidate utterances and form same-prompt chosen--rejected pairs. We then adopt a two-stage optimization strategy: Rejection Sampling Fine-Tuning (RSFT) first adapts the model to LALM-selected high-scoring samples, followed by Anchored Flow-DPO, which formulates pairwise preference optimization using utterance-level flow-matching loss and retains the chosen-sample flow-matching objective as an SFT anchor. This design enables DPO-style preference learning without explicit sequence likelihoods while preserving direct supervision on preferred realizations. On the official 1,600-utterance NVVSpeech Challenge Track~2 test set, our method achieves a Final Track2Score of \textbf{75.80} (79.39 ZH / 72.21 EN), outperforming the VoxCPM2 baseline by \textbf{+1.84}. The improvements are mainly driven by higher NVV Accuracy and NVV Perceptual Effect, while Overall Quality remains stable.
Abstract (translated)
URL
https://arxiv.org/abs/2609.11260
PDF
-
Domain-Incremental Learning for Multi-Channel Replay Speech Detection
Abstract
Replay attacks are the most accessible threat to voice-controlled systems, and the acoustic cues that expose them are strongly modulated by the environment in which the attack is mounted. A detector deployed in the field therefore has to absorb new acoustic conditions over time, ideally without revisiting past recordings, since retaining speech indefinitely is both expensive and legally constrained. We frame this as Domain-Incremental Learning (DIL) over acoustic environments and present the first continual learning benchmark for multi-channel replay speech detection, evaluating a state-of-the-art beamformer-based detector over all 24 environment orderings of the ReMASC corpus with five seeds. Sequential fine-tuning forgets severely, raising the error rate on previously learned environments by 18.8 points. Elastic weight consolidation (EWC) halves forgetting but loses plasticity, gradient projection memory (GPM) is statistically indistinguishable from naive fine-tuning, and the proposed task-specific beamformer (TSB) that keeps one spatial front-end per environment significantly improves final and incremental accuracy. We further show that the last environment of the sequence dominates final performance. Code, results, and analysis are available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2609.11194
PDF
-
The Machines Are Calling: Measuring Automated and Synthetic Voices in Unwanted Inbound Calls
2026-09-10 06:28:50Xingyu Shen, Tommy Duong, Muduo Xu, Xiaodong An, Jiaqi Gan, Haoyuan Tang, Jamey Z. Liang, Siyu Zhang, Yan Zhang, Simiao RenAbstract
In February 2024 the U.S. Federal Communications Commission (FCC) placed AI-generated voices under the Telephone Consumer Protection Act (TCPA). Yet no peer-reviewed measurement says how much unwanted call traffic is placed by a machine, or how much of that machine speech is synthesized rather than played from a recording. We report both with a disclosed pipeline. An interactive voice honeypot (language-model personas on real U.S. numbers, the caller recorded on its own track) recorded 10,987 calls over 66 days; 11 days on which our stack answered silently are set aside. Three instruments read each opening: an audio fingerprint that finds the same recording played on other calls, a commercial synthetic-speech detector on the caller's first ten seconds, and blinded listeners who check what it flags. Of the 7,233 calls our persona greeted on normal days, 13.8% open with a recording we also heard on another call, and 13.1% with fresh audio the detector labels synthetic. A further 9.9% open with a caller who never spoke after our greeting, 54.2% with fresh audio the detector labels human, and 9.0% could not be scored. Machine-voiced openings are therefore at least 26.9%, a further tenth of calls are silent connections we read as machine-placed, and replays of a recording make up 45% of the detector's own rate (29.3% of 6,192 scored openings). The same waveform played on two calls lands on opposite sides of the detector's threshold 13.6% of the time, and eleven listeners confirm 54.4% of what it flags. Synthetic openings concentrate in lead-generation spam (33.8%), not fraud (21.1%); 0.44% disclose automation. Prevalence tracks how long a bait number has circulated (59% against 19% in the same weeks): seeding history, not calendar time, explains the trend. Campaigns outlast their numbers: one recorded compliance notice opens calls in six campaigns, and one synthetic voice serves nine.
Abstract (translated)
URL
https://arxiv.org/abs/2609.11137
PDF
-
Less can be More: What Aspects of Speech Drive End-of-Turn Detection
2026-09-10 04:12:18Rini Sharon, Manickavela A, Kadri Hacioglu, Andreas StolckeAbstract
In conversational AI, detecting when a speaker has finished talking is crucial for natural turn taking. While recent work incorporates semantics, the relative contribution of different modalities remains unclear. We present a controlled ablation of acoustic, prosodic, and semantic signals for streaming end of turn detection using a lightweight trimodal classifier. Under identical training conditions, the acoustic prosodic combination achieves the best balance of accuracy and latency, achieving utterance F1 of 0.93 with 7.8% false alarms at 400ms median latency. Adding text increases premature detections without improving performance. Feature space analysis confirms that prosodic features have the strongest class separability, while text representations overlap substantially. These findings suggest that turn-taking is primarily conveyed through intonation and silence patterns rather than semantic completeness, enabling faster and more reliable systems without expensive text inference.
Abstract (translated)
URL
https://arxiv.org/abs/2609.11066
PDF
