TurnTurk‑GPT is a Turkish text‑based turn‑taking predictor created by fine‑tuning a Turkish GPT‑2 model with the TurnGPT objective on both synthetic and real dialogue. The model is evaluated on real Turkish conversations using word‑level decision points on unpunctuated, lower‑cased text, and it outperforms a multilingual end‑of‑turn detector in both recording‑disjoint and speaker‑disjoint settings. While it matches a zero‑shot Turkish language model on unseen speakers, it only surpasses it when test speakers appear in training; removing verb‑final suffixes significantly degrades performance, highlighting the importance of Turkish morphology for turn detection.
arXiv:2608.22071v1 Announce Type: cross Abstract: Turn-taking is a basic organizational feature of human conversation and remains difficult to model in natural, synchronous dialog systems. While exis...
The study investigates how the speech genre of training data affects end‑of‑turn (EoT) detection. By training a streaming ConvLSTM EoT detector on matched corpora—multi‑turn human‑human conversations versus single‑turn human‑machine interactions—the conversational model achieves up to +0.45 higher F1 scores across four test sets, mainly due to fewer false alarms on mid‑turn pauses and hesitations. The advantage persists at both frame‑level and utterance‑level decoding and is evident only when evaluated on conversational data, not on single‑turn speech.
TurnBench is a new multi‑domain benchmark for evaluating turn‑taking dynamics in spoken dialogue. It comprises a 30‑hour hand‑labeled corpus of dyadic human conversations, a standardized evaluation protocol for end‑of‑turn and interruption detection, and covers six distinct interaction styles with triple annotation. The benchmark also provides a 104‑hour training set, a public leaderboard, and an interactive dataset viewer at https://turnbench.sesame.com.
arXiv:2607. 01345v1 Announce Type: cross Abstract: Turn-taking naturalness is central to full-duplex spoken dialogue systems, yet its automatic evaluation remains limited.
The study investigates which speech aspects best signal the end of a speaker’s turn in conversational AI. By systematically ablating acoustic, prosodic, and semantic cues in a lightweight trimodal classifier, the authors find that combining acoustic and prosodic features yields the highest accuracy and lowest latency, achieving an utterance F1 of 0.93 with 7.8% false alarms at 400 ms median latency. Adding textual information actually increases premature detections without improving performance, and feature analysis shows prosodic cues provide the strongest class separability while text representations overlap significantly.
arXiv:2603.26515v2 Announce Type: replace-cross Abstract: Efficient and robust turn-taking detection remains a significant challenge in industrial-grade voice AI agents. Existing systems often rely o...
TalkFa introduces a unified benchmark for Farsi dialogue generation and understanding, comprising three datasets: WIKI‑FADIAL (4.2K Wikipedia‑grounded dialogues), DAILYDIALOG‑FA (6.6K dialogues with dialogue‑act and emotion annotations), and PLAYDIAL‑FA (2.1K theatrical dialogues with sentiment labels). All dialogues are curated through multi‑stage review by native speakers, ensuring high quality. Experiments show that LoRA fine‑tuning improves generation performance with less data, while specific models excel on classification tasks, and human evaluation confirms the benchmark’s reliability.
arXiv:2610.20327v1 Announce Type: new Abstract: India's 22 scheduled languages are spoken by over a billion people in conversations that mix languages, often within one turn, overlap freely, and take...
The paper introduces ECHO, a paired diagnostic benchmark for evaluating Chinese real‑time spoken dialogue systems on turn‑taking. ECHO pairs examples that share the same overlap transcript but differ in preceding multi‑turn context, requiring either Yield or Keep actions, and also includes off‑talk cases to test unnecessary yielding. Experiments on four speech systems reveal that three systems over‑yield, correctly keeping the floor on fewer than 13% of backchannels, while the fourth system shows a more balanced performance, illustrating that interruption‑only evaluation can overestimate turn‑taking reliability.
BanglaTurn is a new corpus of 35,374 Bangla podcast speech samples, each 3 to 15 seconds long, labeled for end‑of‑turn detection through speaker diarization, an LLM pass, and human verification. A Whisper‑based model with task‑specific classification heads achieves 84.33 % accuracy on a balanced test set, outperforming the Smart‑Turn v3 baseline (69.28 %) and reducing the false‑negative rate from 51.57 % to 7.55 %, though with a higher false‑positive rate. The study also details the contributions of encoder‑layer fine‑tuning, multi‑scale pooling, INT8 quantization, and reports inference latency of 165–191 ms on CPU.
ufakzeka-kulak is an open Turkish speech‑to‑text model with 31.6 M parameters, built using a FastConformer encoder, a token‑and‑duration transducer (TDT), and a CTC head. It was trained from scratch on 4,977 hours of data—281 hours of human transcription and 4,696 hours of machine‑generated subtitles that were verified by a second reading—under licenses permitting commercial use. The model outputs cased, punctuated Turkish text with numbers written as words, achieving a normalised WER of 0.0921 on FLEURS Turkish and 0.0916 on Common Voice 27.0 Turkish, outperforming Whisper base and small models, and it is released under Apache‑2.0.
The paper introduces a trilingual spoken hallucination detection benchmark covering English, Russian, and Kazakh news, with 12,013 samples that include synthetic alterations and severity levels, as well as 290 fact‑checked misinformation items. Detectors are evaluated in a reference‑free setting on text, ASR transcripts, and audio, revealing that most models underperform baseline classifiers, except Gemma‑3n on transcripts. Synthetic‑trained detectors achieve high macro‑F1 scores on real‑world misinformation, but Russian provenance analysis highlights model‑dependent signals that confound synthetic benchmarks.