This Week in Conversational AI Research: June 29–July 05, 2026

We present Wan-Streamer, a native-streaming, end-to-end interactive foundation model designed from the ground up for real-time, low-latency, full-duplex audio-visual interaction. Wan-Streamer seamlessly models language, audio, and video as both input and output within a single Transformer, where the sequence is represented as interleaved visual, audio, and text input tokens together with visual, audio, and text output tokens, coordinated by block-causal attention for incremental streaming. Unlike cascaded interactive systems that rely on separate VAD, ASR, language, TTS, audio-driven animation, or video-generation modules, Wan-Streamer does not rely on external language, speech, avatar, or video-generation modules: perception, reasoning, generation, response timing, turn management, and cross-modal synchronization are learned jointly within one unified model, reducing pipeline latency and error accumulation. To support natural audio-visual responsiveness, we redesign the entire stack around streamability, including causal encoders, causal decoders, block-causal attention, and low-latency multimodal token scheduling, enabling streaming units as short as 160 ms at 25 fps. Wan-Streamer achieves approximately 200 ms model-side response latency and approximately 550 ms total interaction latency when combined with 350 ms bidirectional network latency, supporting sub-second duplex audio-visual communication. These results position Wan-Streamer as a unified, end-to-end, multimodal interactive foundation model for low-latency streaming interaction.

Existing Reinforcement Learning (RL) research for Text-to-Speech (TTS) focuses on large language models (LLMs), leaving Flow-Matching (FM) under-explored. We present FlowTTS-GRPO, an online RL framework for FM-based TTS. By converting ordinary differential equation (ODE) trajectories into stochastic differential equation (SDE) paths, our method enables direct fine-tuning of open-source FM models without auxiliary models. We show that a weighted reward combination converges faster than a probabilistic scheme, and identify three practical optimizations: omitting classifier-free guidance (CFG) during training accelerates convergence; synthesizing hard cases improves robustness; and applying RL to the FM component enhances audio-detail metrics. Experiments on CosyVoice 3.0 and F5-TTS demonstrate objective and subjective preference gains in speaker similarity and perceptual quality, with F5-TTS also improving intelligibility.

Recently, zero-shot text-to-speech (TTS) has enabled high-fidelity and expressive speech synthesis, but it often fails to imitate unseen speaking styles from uncommon scenarios (e.g., crosstalk, dialects). Moreover, fine-tuning pretrained models requires large, high-quality datasets, limiting rapid personalization. We propose VoiceTTA, a reinforcement learning-based test-time adaptation (TTA) method that improves voice imitation of pretrained zero-shot TTS models. VoiceTTA introduces two style rewards based on coefficient-of-variation differences of F0 and energy, combined with speaker similarity and intelligibility (WER from a pretrained Whisper model), and optimizes learnable prefixes via group relative preference optimization (GRPO) in a flow matching-based model at inference time. Extensive experiments demonstrate substantial improvements on uncommon speech prompts, outperforming state-of-the-art baselines. Audio samples are available at https://voicetta.pages.dev/

Recently, Large Language Model (LLM)-based Text-to-Speech (TTS) models have achieved remarkable naturalness. However, the standard Supervised Fine-Tuning paradigm often converges to statistically averaged prosody, limiting emotional expressiveness. While preference-driven optimization offers a promising alternative, existing approaches suffer from two structural mismatches: information conflict, where content and emotion in a shared latent space produce conflicting gradients, leading to reward hacking and semantic degradation; and scale gap, where sparse sentence-level rewards struggle to guide dense frame-level generation. To overcome these challenges, we propose HPRO, a hierarchical progressive reward optimization framework. Within HPRO, we introduce the HD-Emo codec as a novel differentiable reward model to resolve the information conflict. It extracts speech into distinct content and style preference tokens, structurally isolating emotional optimization from semantic content. Building upon this structured preference space, HPRO bridges the scale gap by progressively aligning frame-, word- and sentence-level objectives. Experiments demonstrate that HPRO significantly enhances emotional expressiveness, while effectively preserving linguistic intelligibility. The code and audio samples are publicly available at https://xxh333.github.io/hpro-demo/.

Classifier-free guidance (CFG) is widely used in flow-matching-based zero-shot text-to-speech (TTS), where generation is typically controlled by two conditions: the target text and a prompt speech signal. Standard CFG strengthens these conditions jointly, while recent branch-selective guidance methods attempt to enhance text or speaker conditioning separately, often leading to a trade-off between text correctness and speaker similarity. In this paper, we revisit the CFG under independently masked text and speech-prompt conditions, and decompose the guidance field into text, speaker, and joint residuals. We show that conventional speaker-selective guidance entangles the speaker residual with the joint residual, which may disturb text-related generation. Based on this observation, we propose joint residual reweighting, which independently controls the speaker and joint residuals within the standard CFG framework. Experiments on F5-TTS and CosyVoice2 show that the proposed method improves speaker similarity while maintaining competitive text correctness, demonstrating the usefulness of the joint residual for balancing speaker fidelity and text accuracy in zero-shot TTS.

Unifying speech, sound, and music generation in one model is hindered by tradeoffs between fidelity, end-to-end training, in-context conditioning, and variable-length synthesis that no current paradigm fully resolves. To address this challenge, we present AudioCALM, a universal audio generation framework that extends autoregressive (AR) next-token prediction from discrete tokens to continuous audio latents: a thin flow-matching head replaces the softmax to predict rectified-flow velocities at each position, and a block-causal AR-Flow attention pattern produces arbitrary-length output. Joint training of multiple audio generation tasks faces an asymmetric text--audio mismatch: speech transcripts align to specific time spans and demand tight, time-aligned attention, whereas sound and music captions describe only overall semantics and rely on diffuse, holistic attention; mixing the two disproportionately degrades sound and music generation. We address this asymmetry at two levels: a data reformulation strategy that unifies all three tasks under a single description-style conditioning interface, and a novel architecture Asymmetric Mixture-of-Modality-Experts (A-MoME), which adds a dedicated residual expert for speech while sound and music share the backbone, incurring no inference overhead on non-speech inputs. Experimental results demonstrate that AudioCALM matches modality-specific state-of-the-art and outperforms prior unified baselines on speech, sound, and music generation benchmarks.

We present ZONOS2 8B, our latest TTS model, which achieves state-of-the-art naturalness, prosody, and voice cloning fidelity. We improve upon Zonos-v0.1 across scale, data, and training recipe. We scale the model from 1.6B to 8B total parameters (900M active) with a novel mixture-of-experts (MoE) backbone, improving inference latency and throughput. We expand our training corpus from 200K to over 6M hours using a new data processing pipeline, and we simplify our post-training and conditioning recipes to improve naturalness and voice cloning fidelity. We evaluate ZONOS2 8B on quality, speaker similarity, WER, and ZTTS1-Eval, our novel TTS benchmark, where it performs competitively with state-of-the-art systems while maintaining good streaming latency. We release our model weights and example inference code under an Apache 2.0 license on GitHub and Hugging Face.

Humans tend to speak louder and clearer in challenging environments, such as noisy conditions or when addressing hearingimpaired listeners, which is called Lombard effect. To simulate this behavior in speech synthesis systems, we introduce a flow-matching based text-to-speech (TTS) model trained with vocal effort and articulation pseudo-labels. The proposed model achieves continuous and disentangled control of vocal effort and articulation, while also enabling word-level emphasis for clarifying specific segments of an utterance. Experimental results show that these control mechanisms effectively improve clarityrelated acoustic features. Furthermore, speech-in-noise experiments demonstrate that our model successfully simulates the intelligibility gains of human clear speech in noisy conditions.

Diffusion-based text-to-speech (TTS) models have achieved significant improvements in speech quality. However, modeling sharp prosodic transitions and rapid pitch variations in expressive speech remains challenging. Existing diffusion-based TTS decoders commonly utilize periodic nonlinearities such as Snake activation function to capture harmonic structures, but this activation funcation provides limited adaptability when modeling abrupt amplitude and frequency variations. In this paper, we investigate the role of oscillatory inductive bias in diffusion-based TTS decoders and introduce an adaptive oscillatory nonlinearity that enables controllable periodic modulation while maintaining signal stability through a linear bypass component. We refer the resulting TTS system as OscillaTTS. Experiments on the LJSpeech and Emotional Speech Dataset show consistent improvements across objective and subjective evaluations, indicating improved modeling of expressive prosodic dynamics.

Speech Language Models achieve reasoning capabilities, but are often hindered by massive parameter counts and a tendency to prioritize linguistic priors over acoustic features. While contrastive decoding enhances grounding by contrasting audio-aware and text-only logits, it increases inference latency. We propose Contrastive Audio-Aware Distillation (CAAD), a framework that internalizes the teacher's contrastive reasoning into the student model's weights. To overcome the high computational training overhead in the dual-path token-by-token contrastive distillation process, we introduce a synchronized teacher-forcing strategy. Anchored by unified Pseudo-Ground Truths, this mechanism enables simultaneous full-sequence generation of the teacher's contrastive distributions, allowing student to distill the audio-aware signal efficiently. Overall, CAAD yields a ~8% relative gain over standard knowledge distillation on Dynamic-SUPERB and successfully reduces linguistic bias in MCR-BENCH.

Connecting a pre-trained speech encoder to a Large Language Model (LLM) is the standard architecture for building Speech LLMs. However, a structural misalignment exists between the encoder and the LLM. Unlike encoders based on automatic speech recognition, which often produce representations in separate language-specific spaces, LLMs operate within a unified language-agnostic space. A mechanism is required to align the encoder's language-specific representations with the LLM's shared space. We argue that speech translation provides a principled way to achieve this. Unlike monolingual transcription, translation requires the model to bridge different languages and learn language-agnostic representations. We experimentally evaluate the impact of incorporating translation objectives into speech encoder pre-training. Our results demonstrate that translation-enhanced pre-training improves cross-modal integration and leads to superior performance across downstream Speech LLM tasks.

Audio-Visual Speech Recognition takes two input modalities, acoustic and visual streams, where visual information from lip movements aids recognition when audio is noisy. Recently, LLM-based AVSR models have emerged as a promising paradigm by connecting pre-trained audio-visual encoders to an LLM, achieving strong results in clean conditions. However, these models are predominantly optimized for clean acoustic conditions, with limited attention to making the LLM backbone robust to noise. No explicit mechanism is employed to produce stable representations under corrupted audio, leading to performance degradation in noisy environments. To address this, we propose VIB-AVSR, which integrates Variational Information Bottleneck layers at targeted positions within the LLM backbone to regularize representations. VIB-AVSR reduces degradation under noisy conditions across multiple SNR levels and noise types, without requiring architectural modifications or additional training data.

We study the problem of physically plausible shadow casting when animating 3D Gaussian Splatting (3DGS) avatars, either individually or in multi-avatar and object-interaction scenarios, within existing 3DGS scenes. In contrast to prior methods that rely on binary hit tests and mesh-based shadow casters, our method performs shadow computation entirely in Gaussian space, without requiring any mesh reconstruction. We introduce RAGA, a Ray-Traced Gaussian Shadow Casting formulation based on exact ray-Gaussian line integrals. For each occluding Gaussian, we integrate the opacity profile along the shadow ray and normalize by the theoretical maximum integral, producing a weight that captures how the ray traverses the occluder rather than merely whether an intersection occurred. To reduce temporal variance from clothing deformations in animated avatars, we further introduce an avatar proxy representation that stabilizes shadow casting while preserving visual fidelity. We implement RAGA using custom CUDA kernels integrated with the NVIDIA OptiX framework; as such, our shadow tracer runs at rates of about 50 FPS. We evaluate on single-avatar, multi-avatar, and avatar-object interaction scenarios across multiple datasets, demonstrating substantially improved shadow realism, temporal stability, and scene coherence. Our project page is available at https://miraymen.github.io/raga/.

Speech conveys information through both words and vocal delivery. We evaluate four leading production realtime voice systems-OpenAI's GPT Realtime 2, Google's Gemini 3.1 Flash Live, and Alibaba's Qwen3.5 Omni Plus and Omni Flash-on tasks where the words and the delivery patterns both convey meaningful information. Across three consequential scenarios, all four systems act on the words rather than the voice. They end calls with crying callers who insist nothing is wrong, approve wire transfers authorized in frightened voices, and enroll callers whose agreement is clearly sarcastic. Surprisingly, this is often not a failure of perception. When asked directly, three of the four systems reliably identify the distress, fear, or sarcasm they later ignore when making decisions. We observe a similar pattern when these realtime voice systems estimate accent and age, as their responses frequently follow the biases of the words rather than the acoustic properties of the speaker. We term this disconnect between perception and action the emotional intelligence gap of voice AI. Prompting systems to explicitly attend to vocal delivery improves performance only partially and inconsistently. Our findings show that current realtime voice AI systems often behave as if speech had been reduced to a transcript, suggesting that they should be used with caution in settings where the tone and emotion of delivery convey important information.

Speech-to-speech translation (S2ST) should preserve not only lexical meaning, but also expressive attributes: emotion, scenario style (e.g., news reporting vs. dramatic dialogue), and nonverbal vocalizations (NVs). Moreover, collecting cross-lingual target speech that is both translation-faithful and expressively aligned with the source is difficult at scale, making reference-based evaluation impractical. We introduce STEB (Speech-to-Speech Translation Expressiveness Benchmark), a 32.6-hour Chinese--English benchmark that evaluates both standard dimensions (translation fidelity, speaker similarity, duration alignment) and expressiveness dimensions (emotion, scenario style, NV preservation). For expressiveness evaluation, STEB uses a caption-then-summarize framework that converts speech into structured expressive attributes and compares source and hypothesis attributes with an LLM judge. Human validation shows statistically significant correlations with listener judgments across all expressive dimensions. We evaluate six S2ST systems covering cascaded systems, end-to-end models, and speech large language models. Many systems, especially cascaded ones, achieve strong translation fidelity, but they still struggle with emotion preservation (best: 3.82/5) and NV preservation (best: 2.31/5). These results reveal a gap between semantic transfer and expressive transfer, identifying expressiveness preservation as an open challenge for S2ST. Audio samples are available at https://cmots.github.io/steb.github.io/.

We introduce FiCA, a Feed-forward, instant Gaussian Codec Avatar generation pipeline that creates lifelike avatars from a single portrait image. Generating a photorealistic and drivable avatar from just a single image is significantly challenging due to the limited visual information available to accurately infer the 3D appearance and geometry of human heads. To address this, we develop a novel system that combines human-centric vision foundation models with a diffusion model. This system is designed to fully exploit partial visual observations to generate lifelike human avatars. Our proposed diffusion model learns a generative mapping from these partial observations to complete and authentic 3D mesh reconstruction. Additionally, we introduce a feed-forward mesh refinement network that enhances the fidelity and identity preservation of the generated avatars, eliminating the need for person-specific test-time optimization. By leveraging a universal prior model that decodes a generated mesh into a set of 3D Gaussians, we generate a photorealistic 3D Gaussian avatar, capable of being driven with novel expressions in real-time. Our experiments demonstrate that the avatars generated by our feed-forward approach faithfully represent diverse identities and surpass the visual quality of avatars produced by recent competing methods.

We present GeoFace, a geometry-constrained multi-view diffusion framework for consistent face generation from a single input. % While recent multi-view diffusion models achieve photorealistic synthesis at the per-view level, they lack an explicit mechanism to enforce a shared 3D structure across views, often leading to inconsistent geometry across viewpoints. To address this, GeoFace proposes a unified dual-stream framework for joint generation of multi-view RGB images and 3D face geometry, where the appearance and geometry streams interact through shared attention layers. To encourage the two streams to mutually constrain each other, we introduce a geometry-guided attention alignment loss that supervises the cross-attention between appearance and geometry tokens with 3D-consistent correspondences, enabling the appearance stream to correctly reference pose-invariant geometric cues for robust alignment across viewpoints. Geometry is represented as a canonical UV position map, derived from a FLAME mesh fitted to multi-view observations, serving as a view-invariant shared constraint across all generated views. Experiments on RenderMe-360 and NeRSemble demonstrate that GeoFace consistently outperforms existing methods in both visual quality and cross-view geometric consistency, facilitating more efficient 3D reconstruction.

Reconstructing high-fidelity, relightable 3D avatars from a single in-the-wild image is a challenging ill-posed problem, primarily hindered by the scarcity of high-quality PBR data and the complexity of disentangling illumination from intrinsic materials. In this paper, we present a data-efficient framework that leverages the robust priors of a unified pre-trained diffusion backbone to sequentially address texture completion, delighting, and material decomposition. Unlike existing methods that rely on fragmented pipelines or extensive proprietary datasets, we utilize cascaded Low-Rank Adaptations (LoRAs) to adapt the strong generative prior of the diffusion model for each sub-task in UV space. Specifically, we first employ an Inpainting LoRA to complete missing UV textures caused by occlusion, leveraging the model's semantic understanding to generate semantically and photometrically coherent details. Subsequently, a Light-Homogenization LoRA and a novel Cross-Intrinsic Attention mechanism are introduced to remove baked-in lighting and collaboratively synthesize pixel-aligned PBR maps (Albedo, Normal, Roughness, Specular, and Displacement). To ensure physical plausibility, we impose a UV-space differentiable BRDF shading loss during the decomposition stage, forcing the generative process to adhere to the rendering equation without the artifacts typical of rasterization-based supervision. Extensive experiments demonstrate that our method, trained on fewer than 100 real 3D scans, generates comprehensive, 4K-resolution PBR assets with superior realism and generalization compared to state-of-the-art methods, and all training code and model weights will be released upon acceptance.

High-fidelity and expressive controllable human animation is essential for content creation and digital avatar applications. However, existing methods face a dilemma between expressiveness and disentanglement. Mainstream 2D pose-conditioned approaches suffer from "motion-shape entanglement", leading to the leakage of the driving subject's body shape. Conversely, methods relying on 3D priors (e.g., SMPL) achieve geometric disentanglement but struggle to capture facial expressions and complex gestures, resulting in rigid animations. To this end, we propose EMOSH, a novel framework for high-fidelity controllable human video generation. First, an Expressive Human Model (EHM) is introduced as the core control representation. By explicitly disentangling shape and pose parameters, we fundamentally resolve the body shape leakage issue. Alongside this, a robust motion tracker is designed to accurately estimate EHM parameters from video. Second, we propose a Coarse-to-Fine Hybrid Motion Injection strategy, enabling more fine-grained control over expressions and gestures. Furthermore, we introduce a Spatially-Aligned Conditioning mechanism to bridge the domain gap between training and inference, improving identity consistency. Extensive experiments demonstrate that EMOSH outperforms previous methods in both self-driven and cross-driven scenarios, producing high-fidelity videos with vivid expressions while maintaining shape disentanglement.

Speech-driven 3D facial animation methods face significant challenges in simultaneously achieving high-fidelity motion and precise artistic control at production quality. Existing controllable models typically learn global style control by relying on large-scale, low-quality \emph{in-the-wild} datasets that compromise overall animation realism. Furthermore, these frameworks often lack the fine-grained temporal precision required for demanding tasks such as dialogue localization (e.g., dubbing), where matching specific facial expressions is as critical as lip synchronization. We present KM-Speaker (Keypoint-Matching Speaker), a novel keypoint-conditioned flow-based generative framework that provides both global style guidance and frame-level temporal control from reference performances. We propose a disentanglement strategy that separates audio-driven lip motion from keypoint-driven upper-face dynamics, together with a global style context preservation mechanism to ensure coherent full-face expressiveness. KM-Speaker advances example-based 3D facial animation by achieving high-fidelity motion and flexible controllability in a data-constrained setting, consistently outperforming state-of-the-art methods in lip-sync accuracy, style adherence, and expressive temporal control.

Generating lifelike facial animation for dyadic conversations requires reconciling high-level cognitive intent with precise low-level motor reflexes, yet existing methods fall short in the semantic understanding of dialogue context and in precise dynamic control. In this paper, we propose MindFlow, a dual-pathway generative framework inspired by the Ventral-Dorsal pathway model in neuroscience, which decouples generation into two collaborative streams, thereby harmonizing deep semantic reasoning with fine-grained control. In the Ventral module, we transform the conventional Sentence-Action approach into a novel Chunk-State approach that models raw acoustic streams as a context-aware, evolving emotional state chain, capturing subtle paralinguistic nuances and mid-utterance emotional shifts missed by sentence-level modeling. The Dorsal module features a conditional autoregressive flow matching network for high-fidelity facial motion, driven by high-frequency acoustic cues and modulated by emotion states, plus a Selective Acoustic Injector for adaptive audio gating to ensure robustness in talking-and-listening dynamics without interference. Extensive experiments demonstrate that MindFlow achieves superior semantic appropriateness and motion naturalness compared to state-of-the-art baselines.

Reinforcement learning from human feedback (RLHF) for 3D generation is now established across a number of works, but most existing pipelines optimise explicit surface representations, often by converting radiance fields into meshes and training heavily on surface-supervised data. We instead fine-tune a pretrained 3D-aware generative model directly from a learned reward over radiance-field density ($σ$) values, with no externally supplied mesh or shape prior. The reward model requires no pretraining, trains easily on a small set of preference samples, and yields robust improvement in 3D geometry. Working on an unconditional 3D-aware face GAN (EG3D), our reward reads the continuous 3D density field of the neural radiance field (NeRF) directly and supplies a geometry-only learning signal, requiring neither text conditioning, mesh extraction, nor multi-view rendering. A density-consistency constraint keeps the 2D appearance qualitatively similar while the geometry is reshaped, at a measurable but bounded distributional cost (FID-50k rises from 4.09 to 6.66): the fine-tuned generator, trained from the preferences of a single annotator as a proof of concept, produces face geometries preferred by users in 74.4% of pairwise comparisons.

Uncalibrated volumetric video streaming for human reconstruction is essential for holographic communication and AR/VR, yet remains challenging due to the need for temporal consistency and computational efficiency from sparse-view inputs. Existing methods rely on per-scene optimization or calibrated cameras, while recent feed-forward models are limited to low-resolution (0.5K) single-frame synthesis. We present HiReFF, a feed-forward method for 2K-resolution 360° human video reconstruction from uncalibrated sparse-view videos. Our framework decomposes the problem into two key tasks: foreground 3D Gaussian reconstruction from sparse-view videos (four views separated by 90°) and computationally efficient high-resolution synthesis. To enable the former, we propose Scale-synchronized Camera Calibration to resolve scale ambiguity for multi-view supervision, and Gaussian-wise Foreground Masking to reconstruct clean foregrounds by modulating Gaussian parameters. For efficient high-resolution synthesis, our High-resolution Side-tuning achieves 2K rendering by augmenting the Gaussian head with supplementary features while keeping the backbone at 0.5K, drastically reducing computational overhead. Experiments demonstrate that HiReFF significantly outperforms existing methods in high-resolution streaming volumetric video reconstruction. https://iridescentjiang.github.io/HiReFF

In regulated domains such as banking and healthcare, where privacy constraints make real speech costly to collect and retain, synthetic speech from modern text-to-speech (TTS) is an appealing alternative for training automatic speech recognition (ASR) without exposing sensitive customer recordings. Yet a persistent distributional gap between synthetic and real data limits how far it can replace genuine recordings. Prior work largely treats this gap as a black box to be engineered around, but in our work, we instead examine its origin directly by probing a SLAM-ASR architecture. Then, we localise where its LLM backbone separates real from synthetic speech and find the discriminative signal concentrated in the early-to-middle layers, where temporal and prosodic perturbations disrupt it most. We further show that representation-level separability, help, but does not directly predict downstream ASR gains. On the other hand, convolving synthetic audio with room impulse responses (RIRs) narrows the gap not by making synthetic speech sound cleaner or more natural, but by reproducing the acoustic irregularities of real recordings. Translating these findings into the training procedure, by adding a layer-selection module combined with RIR augmentation matches a fully real-data baseline using only 25% of the real speech (13.6h) and surpasses it at all higher proportions.

Aligner-Encoders are recently proposed seq2seq end-to-end ASR models that replace decoder attention by predicting the uth token directly from the u-th encoder position, so the encoder must learn the alignment internally without cross-attention or a transducer lattice. In practice, this alignment often forms abruptly in the upper layers, making training sensitive and brittle on long utterances. We propose InterAligner, which adds an intermediate Aligner objective so alignment can form progressively across depth, together with an intermediate CTC loss (InterCTC) to stabilize optimization. On LibriSpeech with a 17-layer Conformer, a final-only Aligner reaches 5.0/7.8 WER (test-clean/other). InterCTC improves to 3.4/6.0, and InterAligner further reduces WER to 3.1/5.6 with the largest gains on long utterances.

Non-autoregressive automatic speech recognition (ASR) enables parallel decoding, but many refinement-based methods begin from random, fully masked, or fixed-length token sequences, requiring multiple iterations to reconstruct the complete transcript. We instead formulate ASR decoding as a variable-length edit refinement of a greedy connectionist temporal classification (CTC) hypothesis. An acoustic-conditioned Edit Flow decoder operates directly on the collapsed CTC hypothesis, predicting insertion, deletion, and substitution operations in parallel. The Edit Flow decoder is jointly trained with a CTC model using a continuous-time discrete diffusion loss. During inference, we find that just two edit steps yield substantial Word Error Rate (WER) reductions, and classifier-free guidance (CFG) further enhances recognition quality by focusing the model on audio features. We also constrain edit proposals using CTC confidence to improve accuracy. Finally, ablation studies validate our design choices, while decoder pretraining and pretrained encoder integration yield significant additional performance gains.

Thousands of languages are spoken worldwide, yet many remain under-resourced for Automatic Speech Recognition (ASR) due to the limited availability of high-quality transcribed speech data. Collecting accurate transcriptions is often costly and labor-intensive, particularly for low-resource languages. In this work, we investigate the use of aligned audio-image pairs to adapt pretrained audio encoders without requiring transcription data before supervised fine-tuning. Our proposed representation alignment stage is introduced between large-scale pretraining and supervised ASR fine-tuning. Specifically, image representations extracted from pretrained vision encoders are aligned with audio representations to further adapt a pretrained audio encoder. For this alignment process, we utilize the Vaani dataset, in which images serve as prompts for speech collection, naturally providing paired audio-image data. We evaluate the proposed approach using multiple vision encoders and a pretrained FastConformer audio encoder. Experimental results demonstrate that models fine-tuned after representation alignment consistently achieve improved ASR performance compared to direct fine-tuning. These findings highlight the potential of audio-image representation alignment as an effective transcription-free adaptation strategy for enhancing ASR systems in low-resource language settings.

Large pretrained text-to-speech (TTS) models sound almost human for well-resourced languages, but much worse for languages that are rare in their training data. We study this quality gap for Khmer and Korean using VoxCPM2, a 2.4B-parameter, tokenizer-free TTS model that joins a MiniCPM-4 language-model backbone with a flow-matching diffusion decoder. We build one shared, language-tagged corpus of about 26 hours and adapt VoxCPM2 with a single Low-Rank Adaptation (LoRA) adapter, trained on both languages at once and added to both the language model and the decoder. The adapter is zero-initialized, so training starts exactly at the original (zero-shot) model. In native-speaker listening tests, the Khmer Mean Opinion Score (MOS) rises from 3.85 to 4.23 with the best adapter (rank 64), a highly significant gain (paired Wilcoxon test, p<0.001), while training only 0.19 to 3.03 percent of the parameters. The automatic loss and the human ratings, however, disagree on the best rank: validation loss is lowest at rank 128, yet MOS peaks at rank 64. The same adapter brings no gain for Korean, a language the base model already handles well, and at a high rank it even degrades quality. Adaptation therefore helps mainly where the base model is genuinely weak.

Human motion follows a temporal hierarchical structure, transitioning from low-frequency global trajectories to high-frequency details. Inspired by the success of multi-level autoregressive models in computer vision, we propose MotionMAR, a coarse-to-fine framework for motion reconstruction from sparse observations. It first estimates the global trajectory of human motion and then gradually refines the temporal details. This architecture consists of four integrated components. The Temporal Multi-scale Tokenization (TMT) VQ-VAE encodes the data at multiple temporal resolutions, separating semantic motion from minor jitters. The Motion Autoregressive Network (MAN) operates in this latent space, predicting motion across scales. It first establishes the global structure through coarse indices and then generates finer indices to recover specific details. Meanwhile, the Scale-Aware Control (SAC) module integrates sparse tracking data to ensure the generated output aligns with actual observations. The Motion Refinement Network (MRN) subsequently smooths consecutive poses and eliminates quantization artifacts. Experiments show that MotionMAR achieves state-of-the-art accuracy on the AMASS dataset, providing a reliable and structure-aware approach for motion reconstruction. The source code is publicly available at http://www.lidarhumanmotion.net/motionmar/.

Accent conversion and controllability remain fundamental challenges in cross-lingual text-to-speech (TTS), particularly for low-resource and phonetically diverse Indic languages. While recent large language model (LLM)-based TTS systems exhibit strong cross-lingual generalization, they provide limited explicit control over accent characteristics and intensity. In this paper, we propose CrossAccentTTS, a framework that enables both accent control and conversion while preserving speaker identity. Specifically, we introduce an Accent Intensity Controller (AIC) that injects weighted language embeddings into the accent subspace, allowing smooth interpolation between accents and fine-grained modulation of accent strength at inference time. Experiments on the Indic Multilingual and L2-arctic datasets shows that CrossAccent-TTS achieves precise control of accent intensity, outperforming strong baselines in accent similarity and controllability by maintaining speaker similarity and naturalness.