Speech
-
Preference-Aware Pareto Steering for Multi-Task Speech LLMs
2026-10-19 19:35:51Liuyuan Jiang, Vishal Sunder, Samuel Thomas, Brian Kingsbury, Jeff Kuo, Lisha Chen, George SaonAbstract
Speech large language models (LLMs) perform multiple tasks, including automatic speech recognition (ASR), timestamp prediction (TS), and speech translation (ST). Joint training of speech LLMs creates task trade-offs, yet standard training with fixed loss weights does not generally control the resulting trade-off according to deployment preferences. To address this, we propose Endpoint-Anchored Preference-Aware Sampling Uniformly along the Pareto Front (EA-SURF), an a priori method that adapts scalarization weights so that the resulting task trade-off reflects the desired preference. EA-SURF retains the standard weighted-loss training pipeline, tracking the current model's performance relative to each task-specific reference and adapting the scalarization weights as training proceeds. We theoretically show that EA-SURF can match a desired preference, with the update becoming stationary at the target under checkable regularity conditions. Across six three-objective speech LLM requests, EA-SURF empirically converges toward the requested targets and increases endpoint-inclusive normalized hypervolume by 17.8% over fixed loss weighting.
Abstract (translated)
URL
https://arxiv.org/abs/2610.27200
PDF
-
Beyond Two Tracks: Speaker-Anchored Full-Duplex Dialogue Recovery
2026-10-19 18:09:47Yuxuan Chen, Yang Xiao, Jack Sun, Chak Sun, Brick Li, SamAbstract
Spoken interaction is moving from alternating turns to full-duplex conversation, where an agent listens while speaking. Learning this coordination requires natural dialogue with persistent speaker-resolved tracks, a data requirement that remains difficult to satisfy at scale. Generative recovery offers a route from ordinary recordings, but recovering two voices is not enough: their track assignments can drift as the conversation unfolds. We address this missing link with speaker-anchored routing before diffusion: a 1.84M-parameter module matches the separator's coarse streams to fixed enrollment identities and orders them before a frozen generator reconstructs the speech, without retraining the generator or altering the reconstructed audio. On 25-minute remote conversations, routing cuts wrong-track time from 43.9% to 4.1%, below the 6.2% of matched-enrollment waveform re-identification, with its largest advantage when one participant holds the floor (17.8% versus 28.6%), and raises fixed-track activity F1 from 0.46 to 0.84 while per-chunk reconstruction quality is unchanged. The same router transfers to unseen face-to-face and telephone recordings, and a margin-gated verification step brings their wrong-track time to 0.9-1.7% while embedding 20-36% of the output, where full re-identification embeds all of it. Recovered voices thus become the participant-consistent tracks that full-duplex dialogue modeling needs. Audio examples: this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2610.27109
PDF
-
Evaluating Tibetan Speech Recognition Across Dialects and Adaptation Regimes: A Multi-Granularity Analysis
2026-10-19 09:44:36Jacob Moore, Sheng LiAbstract
Tibetan speech recognition is usually evaluated with character or syllable error rates because the writing system does not mark word boundaries. These metrics can rank systems differently and respond differently to adaptation failures. We combine character error rate, syllable error rate, and segmented word error rate in one evaluation framework. We use dictionary-based, neural, and language-model-based segmentation and compare their boundaries with expert annotations. We evaluate five ASR architectures on Lhasa Tibetan, then test zero-shot transfer and target-corpus adaptation on five additional corpus subsets. We also vary LoRA rank and training-data size to examine adaptation failure. The results show that a lower character error does not always lead to a lower syllable or segmented-word error. These ranking differences persist across corpora. Metric correlations also weaken in some unsuccessful adaptation settings. Higher LoRA ranks improve syllable and segmented-word scores but remain worse than full fine-tuning in the tested range. A standardized phonological representation provides another view of spelling errors. These findings support evaluation at several linguistic levels and motivate better segmentation and adaptation methods for low-resource ASR.
Abstract (translated)
URL
https://arxiv.org/abs/2610.27002
PDF
-
Think Before You Transcribe: Context-Steered Reasoning for Hotword-Aware Speech Recognition
2026-10-19 05:09:28Qihao Wang, Hang Su, Fan Cui, Lichun Fan, Jian Luan, Yue HuAbstract
Contextual automatic speech recognition (ASR) improves rare-entity recognition by supplying candidate hotwords, but retrieved candidates may be incomplete or incorrect. A reliable system must therefore exploit a relevant candidate without allowing unsupported alternatives to override the acoustic evidence. We propose a context-steered reasoning framework that generates a short, audio-grounded caption before transcription. This task-specific representation lets the model assess candidate relevance before making a lexical decision. Supervised fine-tuning establishes the structured caption-to-transcription behavior, while targeted trajectory injection strengthens group relative policy optimization when difficult utterances yield only unsuccessful rollouts. Experiments across diverse context conditions show that the proposed method consistently improves both transcription quality and hotword recognition while remaining robust to irrelevant candidates.
Abstract (translated)
URL
https://arxiv.org/abs/2610.26986
PDF
-
How Far Can the Gemma 4 Speech Encoder Go Without Changing Its Weights? Saudi and Gulf Arabic Speech Recognition with a Frozen and an Adapter-Augmented Encoder
2026-10-18 21:06:44Amr AbdelkhalikAbstract
We study the 305M-parameter audio encoder shared by Gemma 4 and EmbeddingGemma 2 (identical except for its final projection) for Saudi and Gulf Arabic speech recognition, without changing any of its weights. A frozen system trains only a 22.6M-parameter CTC reader over a mix of encoder layers; an adapted system adds 3.1M low-rank adapter parameters, a 41.5M-parameter hybrid CTC/attention reader and joint decoding with a sub-word language model. Against Cohere Transcribe Arabic (2.07B parameters), re-run on identical audio, the frozen system trails in word error rate (WER) but is statistically indistinguishable in character error rate (CER) on five of six human-labelled conditions, while running 3.4 times faster on CPU with greedy decoding. The adapted system is statistically indistinguishable in WER on a public Saudi customer-service call (19.20% against 19.27%) and on clean Emirati speech, is significantly better on telephone-coded Emirati speech, has a significantly lower CER on all six conditions of these sets, and beats a 0.6B recogniser fine-tuned on SADA in WER on all six; it still trails Cohere in WER on long-form Saudi speech. On a blind, human-corrected public call-centre sample it ranks 4 of 15 systems, behind three commercial APIs (significantly only one) and ahead of the other self-hostable systems by point estimate. An optimised joint decoder runs it at 139 times real time on one GPU. We also show that references made by correcting machine drafts favour the drafting system. Code and public-set results are released.
Abstract (translated)
URL
https://arxiv.org/abs/2610.28673
PDF
-
ShlokaDhara: A Multi-Speaker Sanskrit Speech Corpus for Learning Meter-Aware Prosody in Text-to-Recitation
2026-10-17 14:40:38Harshal Kausadikar, Onkar Kishor Susladkar, Prathmesh Santosh Choudhari, Vedangi Deepak Deshpande, Ismini Lorentzu, Sparsh MittalAbstract
Sanskrit verse recitation is governed by $\textit{Chandas}$, a strict prosodic framework defined by rhythmic $\textit{Laghu}$ (light) and $\textit{Guru}$ (heavy) syllables. Because existing Sanskrit speech datasets prioritize ASR or conversational TTS, they lack the metrical fidelity required for authentic verse synthesis. We introduce $\textbf{\textit{ShlokaDhara}}$, a 43-hour, multi-speaker Sanskrit speech corpus capturing diverse meters, recitation styles, and melodic variations. We specifically focus on $\textit{Text-to-Recite}$, a synthesis task explicitly requiring both prosodic accuracy and metrical correctness. Benchmarking state-of-the-art TTS architectures via an automated UTMOS (neural MOS) predictor demonstrates that XTTS-v2 achieves the highest perceptual quality. We further validate the XTTS-v2 generated samples using two paradigms: subjective assessments by 16 Sanskrit experts and novel empirical metrics, specifically, syllable-level $\textit{Laghu-Guru}$ duration analysis to objectively quantify structural metrical awareness. Ultimately, this work establishes a robust dataset and evaluation benchmark for prosody-aware computational Sanskrit recitation. Generated Samples: this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2610.26929
PDF
-
MedPAO-Assist: Edge-Deployable, Modality-Aware Agentic AI for Radiology Report Structuring
2026-10-17 06:13:06Shrish Shrinath Vaidya, Minmini Selvam, Ganapathy KrishnamurthiAbstract
The adoption of Large Language Models for radiology report structuring has been largely dominated by either proprietary cloud-based models or large open models requiring high-end GPUs for deployment, which is a critical bottleneck for clinical adoption in radiology workflows that demand data privacy and offline operation. To address this, we present MedPAO-Assist, an edge-deployable agentic framework that generates structured radiology reports directly from radiologist dictation and voice commands, requiring only $\approx3.3$ GB peak RAM and achieving an average end-to-end latency of $\approx31$ seconds on a standard laptop CPU through an asynchronous, multi-threaded execution pipeline. The system currently supports four major imaging modalities -- Chest X-ray, Chest CT, Abdomen CT, and Brain MRI. To our knowledge, it is among the earliest agentic systems to run its entire pipeline with a Small Language Model (SLM) in the 0.6B-parameter range. This multi-agent architecture, built around an Extract-Ground-Generate paradigm, performs concept extraction, SNOMED-CT grounding, and structured report generation, followed by a speech-driven human-in-the-loop correction phase, together forming a fully "no-touch" reporting workflow that requires no keyboard input, external API calls, or internet connectivity. We evaluate each pipeline stage individually -- concept extraction, ontology grounding, and concept categorization and report hardware profiling results demonstrating the system's feasibility for real-time, CPU-only clinical deployment.
Abstract (translated)
URL
https://arxiv.org/abs/2610.26913
PDF
-
Jarvis: Coordination Deserves a Brain of Its Own in Agentic Interactive Models
2026-10-16 17:29:10Yuxiang Wang, Shengbo Cai, Zhixian Zhao, Yingda Shen, Yuancheng Wang, Xueyao Zhang, Liqiang Zhang, Teddy Sun, Steve Yves, Zhizheng WuAbstract
Interactive models increasingly hand complex reasoning and tool use to a slow brain in the background, leaving a realtime front model free to sustain full-duplex dialogue. Yet the user may refine, cancel or replace a request while its task runs, so a plan or result that made sense a moment ago goes stale. Each new utterance must therefore be read against the work in flight, but the front model follows only the conversation and the back end sees only its own task. Coordination deserves a brain of its own, one that sees both. We introduce Jarvis, a coordination harness where models judge and code enforces. Its typed decision model, JEV, reads each turn against the live tasks and picks among operations built from a versioned task record. Code checks every choice against the current state before applying it, so no decision can act on a task version the user has replaced. Jarvis also rechecks plans before they execute and results before they are spoken, and counts an answer as delivered only once the user has heard it in full. We build two instances of this design. Jarvis-Audio runs unchanged on three full-duplex speech models. Jarvis-Omni adds vision and rechecks the evidence behind an answer before and during playback, since a scene can change while nobody speaks. With the front model and back end held fixed, replacing Qwen Audio Agent with Jarvis raises tau-Voice task success more than sixfold, from 29 to 192 of 278 tasks, and the gains hold on every benchmark and front model. Against published systems, Jarvis-Audio reaches 69% Pass1 on Full-Duplex-Bench v3 with an open 9B front model, versus 60% for GPT-Realtime. These results suggest that how the two brains are coordinated matters as much as how strong they are.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23804
PDF
-
VocalGrad: Evaluating Acoustic Perception in Audio Language Models
2026-10-16 16:54:35Koki Ryu, Hitomi YanakaAbstract
Recent large audio language models (LALMs) have extended their capabilities beyond lexical speech understanding to tasks requiring paralinguistic and non-linguistic audio understanding, such as emotion recognition. However, these models still struggle to reliably utilize acoustic information, particularly when it co-occurs with spoken linguistic content. Existing benchmarks do not clearly distinguish acoustic perception from higher-level reasoning, making it difficult to identify the source of these failures. To address this issue, we introduce VocalGrad, a benchmark designed to evaluate LALMs' ability to perceive low-level acoustic variations. VocalGrad formulates this problem as a simple binary classification of temporal change direction, enabling evaluation largely independent of higher-level reasoning or semantic understanding. Our benchmark results reveal a substantial gap in acoustic perception between human annotators and state-of-the-art LALMs: while human annotators achieve high accuracy, LALMs perform at near-chance levels. We further conduct linear probing to identify structural bottlenecks in these models. Our analysis reveals two distinct failure modes: for some categories, task-relevant information is preserved in internal representations but not utilized by the language model head; for others, the information becomes progressively less linearly decodable across model layers. Overall, our results reveal substantial limitations of LALMs in modeling acoustic variation under controlled settings and expose representation differences not observable from text outputs.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23745
PDF
-
Beyond the Benchmark: Conditioning, Domains and Failure Modes of Qwen3-TTS
2026-10-16 16:09:20S\'everin Baroudi, Yanis Labrak, Petr Motlicek, Herv\'e Bredin, Ricard MarxerAbstract
Qwen3-TTS reports state-of-the-art zero-shot voice cloning, but its technical report leaves open how the model should be conditioned, how it behaves outside read speech, and how it fails. We characterise its two cloning interfaces, in-context learning (ICL) and X-Vector, against Chatterbox, XTTS-v2 and F5-TTS on read, meeting, emotional and air-traffic-control (ATC) speech, using eighteen objective measures, an LLM judge, a spoofing countermeasure and a listening test. Neither interface dominates: ICL clones timbre more closely from correctly transcribed references, X-Vector clones timing, and ICL degrades sharply when a short reference transcript is wrong. Domain explains four times more variance than system choice. Qwen3-TTS leads on intelligibility and pitch register but trails on prosody, and the objective measures rank systems differently from listeners. An off-the-shelf countermeasure flags its clones no more often than genuine recordings, and on ATC radio it fails for every system. Code and audio are released.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23678
PDF
-
LiveSync: Iterative Self-Training for Lip-Sync Generation
2026-10-16 14:46:38Huaishao Luo, Huan Zhang, Guanbo Bao, Yinglu Liu, Youzheng Wu, Xiaodong HeAbstract
Audio-driven lip synchronization aims to match target speech through localized mouth and jaw articulation while preserving subject identity, head pose, and surrounding facial context. Naturally recorded videos provide synchronized reconstruction targets but rarely offer pairs that are matched in identity and pose yet differ primarily in speech-related articulation. Mask-based training enables reconstruction from these videos, but random-reference inputs differ from the aligned self-reference used at inference. We present LiveSync, an iterative self-training framework in which a single mask-based model generates pseudo pairs and is refined for deployment. The model is first trained on a curated real-video corpus augmented with externally generated clips covering under-represented visual conditions. A frozen checkpoint then uses independently sampled audio to generate mouth-edited references from real-video frames, copying pixels outside the edit mask from each source frame. After confidence filtering, these references are paired with aligned masked inputs and original real-frame targets. The retained pseudo pairs are mixed with real training data to refine the same model from its current weights, updating training supervision without adding inference stages. Although we instantiate the framework with a single-pass GAN-based model, the self-training strategy could in principle also be adapted to diffusion-based models. Experimental results show that LiveSync achieves leading synchronization performance on HDTF and competitive performance compared with the baselines on the AIGC-LipSync subset, while maintaining low inference latency.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23558
PDF
-
ConvexLLM: Scaling Convex-Hull Speech Interfaces to General-Purpose Audio Intelligence
2026-10-16 14:35:04Ming-Hao Hsu, Shujie Liu, Yuxuan Hu, Yuxiang Wang, Yingda Shen, Jinyu Li, Zhizheng WuAbstract
Speech and audio language models connect continuous acoustic representations to decoders pretrained primarily on text data. Most such models use MLP or Q-Former interfaces to map acoustic features into the decoder's input space, requiring the decoder to learn how to interpret the resulting representations. We investigate whether grounding these representations in the decoder's existing text embeddings can help it use acoustic information while better preserving its pretrained text capabilities. We present ConvexLLM, a general-purpose speech and audio language model whose C-Gate interface represents audio as continuous convex combinations of frozen text embeddings. ConvexLLM achieves an eleven-task VoiceBench aggregate of 75.36, the highest among the open-weight models we compare, and an MMAU Test-mini accuracy of 62.50%. Its largest margins over the baselines are on spoken knowledge and reasoning tasks, while the strongest baselines are more accurate on audio understanding and transcription. In a comparison with a 2B-parameter decoder, C-Gate achieves higher aggregate spoken-language performance than MLP and Q-Former interfaces, with text-task performance close to MLP and task-dependent tradeoffs in transcription and audio understanding. These results support convex combinations of text embeddings as a viable interface for general-purpose speech and audio language models.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23539
PDF
-
Interpreting SSL Representations for Spoof Detection: a WavLM Study
2026-10-16 14:26:55Mohamed Mallat, Michele Panariello, Massimiliano Todisco, Nicholas Evans, Anthony LarcherAbstract
Self-supervised speech models such as WavLM achieve strong spoofing detection performance, yet which layers encode the most spoof-relevant information and what acoustic properties they capture remain opaque. We present a layer-wise interpretability study of WavLM using the ASVspoof 5 database, combining three complementary analyses: (i) per-layer linear probing with attack-level evaluation across speech synthesis, voice conversion, and adversarial attacks; (ii) a canonical correlation analysis (CCA) linking layer representations to spoofing-relevant acoustic features; (iii) an exhaustive grid search over layer subsets showing that use of five layers outperforms full thirteen-layer pooling. CCA analysis reveals that, while most acoustic features peak in alignment with mid-layers and then decay, voice quality measures jitter, shimmer, and the harmonics-to-noise ratio show the opposite trend, with correlation increasing toward later layers. Embedding visualisations reveal that deeper layers capture a continuous acoustic similarity among attacks rather than discrete categories, suggesting these layers encode fundamental synthesis artifacts rather than attack-specific signatures.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23526
PDF
-
A low-dimensional dynamical basis for tongue shape control
Abstract
Speech production involves the control of a high-dimensional articulatory system with many degrees-of-freedom, but a widespread hypothesis is that speakers navigate this complexity using a set of lower-dimensional control variables. Common candidates for these control variables include individual articulators (e.g. tongue body) and abstract tract variables (e.g. tongue body constriction location), but identifying the most appropriate coordinate system for articulatory control remains an open question. In this study, we advance a model in which holistic tongue shape modes represent a potential low-dimensional control basis for a simple dynamical law. We first discover empirical tongue shape modes in a cross-speaker coordinate space and then fit the model to ultrasound data on monophthongal vowels from 36 speakers of British English. A three-dimensional linear point attractor fitted to these time-varying modes recovers the dynamics of tongue shape with a high degree of accuracy. This suggests that complex tongue movement can arise from independent modes that linearly combine under simple dynamical laws, thereby relocating some complexity from the control signal to the coordinate system. We show that this framing explains why one-dimensional articulatory measures can exhibit observable nonlinearities, and hypothesize that shape modes are a plausible kinematic expression of underlying muscle synergies.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23456
PDF
-
Flash RNN-T: Memory-Efficient Joint Network and Transducer Loss Computation
Abstract
The recurrent neural network transducer (RNN-T) remains a practical choice for streaming speech recognition, but training requires large intermediate tensors. We present Flash RNN-T, a joint and loss implementation combining specialized kernels with optional tiled recomputation over the full alignment space and vocabulary. Tiled and untiled Flash form the entire measured throughput-memory frontier on medium and long workloads. On H100, untiled Flash achieves 24-39% higher throughput with 26-56% less memory than the fastest baseline on these workloads wherever it fits. Tiling reduces memory further through recomputation. On medium, tiled Flash uses ninefold less memory than that baseline at 16% lower throughput when both use one backward call per batch. On long, it fits where all tested alternatives with that backward schedule run out of memory with 80 GB. These comparisons use each implementation's lowest-memory configuration reaching at least 95% of its peak throughput.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23409
PDF
-
Do Speech BCIs Need Larger Models? Rethinking Neural Decoding beyond Scaling
2026-10-16 09:12:58Zehui Feng, Weichuan Wang, Xiaohan Chen, Cuntai Guan, Ting HanAbstract
Recent speech brain-computer interfaces (BCIs) increasingly rely on large models and complex multi-stage pipelines for neural decoding. However, in speech BCI, where training data are scarce, noisy, and expensive to collect, the benefits of model scaling remain unclear. This work revisits a fundamental question: are large models truly necessary for accurate neural decoding? We propose BEST (Brain-Encoding-Speech-Text), a two-stage framework for efficient neural-to-text decoding. Instead of relying on large generative models or phoneme-centric pipelines, BEST introduces hierarchical intermediate representations and a compact speech-oriented decoder that directly maps neural activity to text. This design preserves essential structure while significantly reducing computational cost, enabling robust generalization in low-resource settings and practical deployment under constrained resources. On the Brain-to-Text '24 and '25 benchmarks, BEST achieves word error rates (WERs) of 6.19% and 1.62%, respectively, using only 3.85% of the parameters required by comparable methods. Beyond these results, we provide three key insights: (1) decoder scaling yields non-monotonic gains; (2) cross-modal alignment is more critical than model size; and (3) overfitting degards sequence-level performance. Representation analysis further reveals a consistent neural-audio-text alignment pathway.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23175
PDF
-
Advancing Sentence-Level Dysarthric Audio-Visual Speech Recognition via Test-Time Adaptation
2026-10-16 08:26:55Shiyao Wang, Hui Wang, Jiaming Zhou, Shiwan Zhao, Yong QinAbstract
Dysarthric audio-visual speech recognition (DAVSR) enhances performance by integrating complementary visual cues. However, existing research is largely confined to word-level tasks with limited vocabularies, which severely restricts real-world communication for dysarthric speakers. Furthermore, performance remains heavily dependent on supervised fine-tuning with speaker-specific data; significant variations in age, etiology, and severity cause traditional methods to struggle with unseen speakers. To address these domain shifts and the scarcity of continuous multimodal resources, we first introduce a "source speech search-segmentation" pipeline to construct a structured sentence-level corpus from raw CDSD videos. We then propose the use of unsupervised Test-Time Adaptation (TTA) for sentence-level DAVSR using Sequential-level Generalized Entropy Minimization (SGEM). Our evaluations across the state-of-the-art AV-SUPERB and ViSpeR frameworks demonstrate that TTA consistently alleviates Character Error Rate (CER) degradation for unseen pathological speakers. Paired t-tests verify that these improvements are statistically significant (p < 0.05), highlighting TTA as a promising paradigm for advancing speaker-adaptive assistive speech interfaces.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23117
PDF
-
X-MDD: LLM-Based Mispronunciation Detection and Diagnosis for Unseen Languages with Canonical References
2026-10-16 07:50:00Xintong Wang, Junchuan Zhao, Ye WangAbstract
Mispronunciation detection and diagnosis (MDD) provides phone-level feedback to language learners, but extending it to new languages typically requires collecting and annotating learner speech for each target language. This repeated annotation effort is costly and limits the language coverage of pronunciation feedback. We study whether MDD capabilities learned from annotated learner speech can be applied to languages unseen during MDD training when canonical pronunciation references are available. Our multilingual LLM-based model, X-MDD, uses canonical speech and phones as references to predict the phone sequence actually produced by a learner. It incorporates in-context demonstrations for inference-time adaptation and is further optimized with group relative policy optimization (GRPO) using detection F1 and phone error rate. We further analyze how canonical inputs affect prediction and find that (1) canonical speech and phones jointly contribute to actual-phone prediction, and (2) conflicting speech evidence reduces the tendency to copy canonical phones. Trained only on Mandarin and English, X-MDD achieves a detection F1 of 38.07 on Arabic Qur'anic recitation, compared with 44.14 for a supervised challenge baseline trained on Arabic. It also achieves F1 scores of 78.68 and 58.78 on Mandarin and English, respectively, demonstrating effective multilingual MDD within a single model.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23079
PDF
-
DriftTTS: One-stage Single Step Text-to-Speech via Generative Drifting
2026-10-16 05:31:38Jaehyun Ahn, Dohyun Kim, Heeseung KimAbstract
Text-to-speech models built on score matching and flow matching produce natural speech, but only by calling the decoder many times, which makes synthesis slow. Existing ways to cut the number of calls either distill from a multi-step teacher or add a stage on top of a pretrained model, leaving single-step TTS without a pretrained TTS model unexplored. We question whether pretraining a TTS model is necessary, and propose DriftTTS, which adopts generative drifting, an objective that moves iterative sampling from inference into training. Drifting assumes samples of fixed shape, whereas speech has variable length; we address this issue with segment drifting. On LJSpeech, DriftTTS matches the multi-step quality of Matcha-TTS and RapFlow-TTS in a single decoder call, while being faster. To our knowledge, this is the first application of generative drifting to text-to-speech.
Abstract (translated)
URL
https://arxiv.org/abs/2610.22960
PDF
-
ASR-Style Transcript Normalization Removes Useful Pragmatic Signal for Dialogue Act Classification
2026-10-16 04:51:37Lucky Onyekwelu-Udoka, Md Shafiqul IslamAbstract
Automatic speech recognition (ASR) pipelines often normalize transcripts by removing filled pauses and punctuation and expanding colloquial forms, potentially discarding cues useful for dialogue act classification. We evaluate nine transcript conditions across six classifiers (TF-IDF logistic regression, RoBERTa, BERT, ALBERT, GPT-2, and BiLSTM) and three conversational corpora (Switchboard, ICSI MRDA, and SLUE-HVB), supplemented by multi-seed experiments, normalization controls, and real-ASR evaluations. Across five random seeds, heavy normalization reduces Macro-F1 in all 15 neural model-corpus combinations under the original preprocessing protocols. Removal using the original filler and backchannel lexicon consistently degrades performance on MRDA and SLUE-HVB but produces mixed effects on Switchboard. Punctuation-aware TF-IDF experiments show that punctuation removal reduces Macro-F1 by 0.036 on Switchboard and 0.078 on MRDA, demonstrating that the original linear baseline's apparent insensitivity resulted from tokenization. Restricting filler removal to um, uh, and erm preserves degradation on MRDA but largely eliminates the SLUE-HVB effect, where the original lexicon removes class-defining backchannel expressions. These findings demonstrate that normalization effects depend on preprocessing definitions, feature extraction, and conversational context, supporting task-dependent rather than indiscriminate transcript normalization.
Abstract (translated)
URL
https://arxiv.org/abs/2610.22934
PDF
