Speech Synthesis, Recognition, and More With SpeechT5

Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster

Introducing gpt-realtime and Realtime API updates

We’re releasing a more advanced speech-to-speech model and new API capabilities including MCP server support, image input, and SIP phone calling support.

Phoneme-guided TTS augmentation for ASR: A unified pipeline and multilingual evaluation

The paper introduces a phoneme-guided text-to-speech (TTS) augmentation pipeline for automatic speech recognition (ASR) that links multilingual speech generation with candidate-text selection and reference-speech quality control. It proposes phoneme-frequency-guided selection (PFGS), which prioritizes candidate texts containing common phonetic content based on real ASR training transcripts. Experiments across four languages and 13 test sets show that random text selection improves recognition on 11 test sets, while PFGS further improves nine test sets with relative word error rate reductions up to 19.3%, and reference-speech filtering also contributes to performance gains.

A Comprehensive Study of Content Representations for Speech Synthesis

The paper investigates how different speech content representations—such as SSL features, supervised tokens, posteriorgrams, and neural audio codecs—perform when used to train a generative model that produces audio conditioned only on each representation. By evaluating the generated audio on content, speaker identity, and prosody, the study identifies two regimes: some representations almost fully reconstruct the original audio, while others effectively separate speaker identity. The findings reveal that disentanglement of speaker identity depends on both the training objective and the representation’s information capacity, rather than supervision alone.

Pronunciation Editing for Finnish Speech using Phonetic Posteriorgrams

arXiv:2507.02115v3 Announce Type: cross Abstract: Synthesizing second-language (L2) speech is potentially highly valued for L2 language learning experience and feedback. However, due to the lack of L...

Intelligent transcription with Gemini 3.5 Transcribe

DeepMind has introduced Gemini 3.5 Transcribe, a new tool that offers more intelligent speech-to-text transcription. The update promises improved accuracy and smarter handling of spoken content, enhancing the overall transcription experience.

CookVoice: Unified Framework for Style Controllable Multi-Modal Human Voice Generation

arXiv:2608. 11590v1 Announce Type: cross Abstract: Human voice generation has made rapid progress in speech generation, singing voice generation, voice cloning, and voice editing.

PEDBench: Can Speech Models Detect Phonetic Errors?

PEDBench is a new benchmark for phonetic error detection (PED) that includes natural recordings of children’s speech errors, scripted error productions by actors, and synthetic error productions generated via phone‑conditioned text‑to‑speech. The benchmark provides high‑quality phonetic annotations verified by trained annotators. Evaluation of 13 state‑of‑the‑art speech models shows that they struggle to reliably detect phonetic errors, even in controlled conditions, and often misclassify correct phonemes as errors. Fine‑tuned speech‑language models perform better on word‑level PED, but a sim‑to‑real domain gap persists.

Phoneme-Guided Initialization for LLM-based Speech Recognition

The paper introduces phoneme‑guided initialization for large language model‑based speech recognition. By pre‑training an audio encoder on speech‑to‑phoneme conversion and an LLM on phoneme‑to‑grapheme conversion, then fine‑tuning jointly on ASR, the method leverages phoneme‑mediated processing within an end‑to‑end framework. Experiments on Japanese, Chinese, Tatar, and Urdu demonstrate that this approach matches or surpasses both cascaded S2P‑P2G baselines and end‑to‑end models without phoneme initialization.

HuPER: A Human-Inspired Framework for Phonetic Perception

HuPER is a human-inspired framework that models phonetic perception as adaptive inference over acoustic‑phonetics evidence and linguistic knowledge. Using only 100 hours of training data, it achieves state‑of‑the‑art phonetic error rates on five English benchmarks and demonstrates strong zero‑shot transfer to 95 unseen languages. It uniquely enables adaptive, multi‑path phonetic perception across diverse acoustic conditions, and all training data, models, and code are open‑sourced.

An Empirical Study on Learning Latent Representations for Emotional Speech Synthesis

arXiv:2606. 14922v1 Announce Type: cross Abstract: For the last couple of years, the field of speech synthesis has improved dramatically thanks to deep learning.

Transsion's Speaker-Attributed Multilingual ASR System for the MLC-SLM 2026 Challenge

The paper describes Transsion Speech Team’s submission to Task 1 of the MLC‑SLM 2026 Challenge, aiming at speaker‑attributed transcription for multilingual conversational speech. Their cascaded framework includes a DiariZen‑based speaker diarization module, a Qwen3‑Omni‑based long‑form multilingual ASR module with CTC alignment for precise timestamps, and a fusion module that merges diarization and transcription outputs into speaker‑attributed STM results. On the official evaluation set, the system achieved a tcpMER of 15.41% and secured second place among all participants.