---
title: Hybrid Spoofed Audio Dataset (HSAD)
url: https://www.emergentmind.com/topics/hybrid-spoofed-audio-dataset-hsad
type: topic
---

# Hybrid Spoofed Audio Dataset (HSAD)

Searching arXiv for the relevant papers and closely related dataset work.
arXiv search query: "Hybrid Spoofed Audio Dataset HSAD audio spoof detection hybrid"
Hybrid Spoofed Audio Dataset (HSAD) denotes a line of spoofed-audio benchmarks organized around the observation that realistic attacks are often not fully synthetic utterances. In the literature, the term is used explicitly for a benchmark containing human, cloned, zero-shot AI-generated, and mixed human–synthetic speech, and more broadly for hybrid corpora that combine multiple spoof types, sources, or annotation layers such as replay conditions or linguistically informed labels [2505.15136][2509.07323][2409.06033]. The unifying idea is that spoof detection should be evaluated beyond the conventional binary setting in which an entire file is either bona fide or wholly fake.

## 1. Terminology and conceptual scope

The acronym **HSAD** is not uniformly used across the literature. One 2024 study consistently refers only to **“a hybrid dataset”** or **“our hybrid dataset”**, describing it as “a hybrid dataset comprised of multiple types of spoofed audio augmented with sociolinguistic annotations,” but it does not introduce the name *Hybrid Spoofed Audio Dataset (HSAD)* or any HSAD acronym [2409.06033]. By contrast, later 2025 work explicitly introduces and uses **Hybrid Spoofed Audio Dataset (HSAD)** as the formal name of a benchmark for mixed AI–human speech [2505.15136][2509.07323].

The term **“hybrid”** itself is used in several technically distinct senses. In the named HSAD work, it primarily denotes utterances that mix genuine human segments with cloned or AI-generated segments inside one file, together with parallel clean and degraded conditions [2505.15136][2509.07323]. In the 2024 linguistically annotated hybrid dataset, “hybrid” refers to a corpus that combines **replay attacks, mimicry, Text-to-Speech (TTS) deepfakes, Voice Conversion (VC) deepfakes, and genuine speech**, while also adding **Expert Defined Linguistic Features (EDLFs)** [2409.06033]. Related research extends the same idea to **partially spoofed utterances** that mix bona fide and spoofed segments [2104.02518], **replay-aware corpora** containing bona fide, replayed bona fide, fake, and replayed fake speech [2510.19414], and **component-level** mixtures where speech and environmental sound can be spoofed independently [2509.15804].

This suggests that HSAD is best understood not as a single immutable dataset specification, but as a benchmark paradigm for **mixed-source, multi-condition, or partially manipulated audio** in anti-spoofing research.

## 2. Canonical HSAD formulation in 2025 work

The most explicit HSAD formulation is given in 2025 work that constructs a dataset around **12 participants**, **ages 19–38**, with **gender balanced** recording, in an **acoustically controlled indoor environment** using **studio-grade USB microphones**, recorded at **44.1 kHz**, then **downsampled to 16 kHz** [2505.15136]. Each speaker records **104 scripted sentences across 8 categories**: **Alphanumeric combinations (8)**, **Pure alphabetic sentences (8)**, **Numeric sequences (8)**, **Natural English sentences (16)**, **Semantically coherent sentence pairs (16)**, **Semantically unrelated sentences (16)**, **Sentences with grammatical errors (16)**, and **Complex sentences with semantic + grammatical anomalies (16)** [2505.15136].

The core HSAD class structure comprises **four balanced classes**: **Human**, **Cloned**, **AI-generated**, and **Hybrid**, with the **Clean Hybrid Dataset** containing **1,248 utterances total**, namely **312 utterances per class** [2505.15136]. A parallel **Noisy/Degraded Dataset** contains **another 1,248 utterances**, obtained by creating **one distorted version per clean utterance** [2505.15136]. A later paper on HSAD reports the same four classes and the same **1,248 clean utterances**, but its abstract also states **“1,248 clean and 41,044 degraded utterances,”** while its detailed construction section still describes a **1,248-clean + 1,248-degraded setup** and does not reconcile the larger degraded count [2509.07323]. This discrepancy is one of the clearest textual ambiguities surrounding the named HSAD.

The synthetic portion of HSAD is divided into two principal categories. **Cloned speech** is generated with a **Tacotron 2-based TTS system plus speaker embeddings**, using four cloning fidelity levels: **\(C_1\)** for a single sentence reference, **\(C_2\)** for a 16-sentence subset, **\(C_3\)** for the full 104-sentence speaker corpus, and **\(C_4\)** for the target sentence embedding [2505.15136]. **Pure AI-generated speech** is produced with **zero-shot TTS systems without explicit speaker identity preservation** [2505.15136].

HSAD’s defining novelty is the **hybrid utterance**, constructed by **segment-level concatenation** under three composition patterns: **H→S**, **S→H**, and **H↔S↔H**, where transitions are smoothed with **10 ms cross-fading** [2505.15136]. Later analytical work on HSAD retains the same core idea and further refines evaluation with a six-group taxonomy: **G1 – Genuine Human**, **G2 – Pure AI Clone**, **G3 – Pure AI Generated**, **G4 – Mixed: AI Generated + Human**, **G5 – Mixed: AI Clone + AI Generated**, and **G6 – Human Recombined** [2505.15136][2509.07323].

## 3. Annotation, metadata, and protocol design

HSAD is structured as more than a four-way label set. Each utterance is annotated with **speaker ID**, **gender**, **spoofing method**, and, for mixed examples, **segment arrangement** such as **H→S**, **S→H**, or **H↔S↔H** [2505.15136]. In the 2025 evaluation-oriented HSAD description, hybrid examples are also annotated with **segment start and end boundaries**, **source type per segment**, and **degradation metadata** [2509.07323]. At the same time, both papers emphasize that the benchmark is still used primarily for **utterance-level classification**, and that **segment-level timestamps are a planned extension** rather than a fully realized localization protocol [2505.15136][2509.07323].

The degraded partition is produced through three explicit perturbation families. First, **additive noise** is injected using noise types such as **street**, **café ambiance**, and **white noise**, at **10, 15, 20, 30 dB** SNR [2505.15136]. Second, **channel simulation** includes **low-pass filtering at 4 kHz** [2505.15136]. Third, **compression artifacts** are introduced via **Opus** at **16 kbps and 24 kbps** [2505.15136]. Each distorted utterance inherits the original class label and additionally carries **SNR level**, **noise type**, and **degradation parameters** such as codec and filter settings [2505.15136].

The main experimental protocol is **speaker-disjoint**. The papers describe an **80/20 train–test split**, with **10% of training** held out for validation, so that speakers in test are different from speakers in train [2505.15136][2509.07323]. This design is intended to suppress trivial voice memorization and to emphasize cross-speaker generalization under hybrid tampering. A later HSAD paper also states that the test set is subdivided into **seen and unseen spoof types** for cross-attack generalization, though it does not provide exact per-split counts beyond the 80/20 ratio and speaker disjointness [2509.07323].

A broader annotation tradition around hybrid spoof data appears in linguistically augmented corpora. The 2024 unnamed hybrid dataset and the later **Linguistically Augmented Audio Speech Data (LinguAS)** add five **Expert Defined Linguistic Features (EDLFs)**—**pitch**, **pause**, **word-initial and word-final release bursts of consonant stops**, **audible intake or outtake of breath**, and **overall audio quality**—encoded as binary indicators per clip [2409.06033][2606.10246]. In that line of work, hybrid spoof datasets are not only mixed by source type but also enriched with human-discernible linguistic metadata.

## 4. Benchmarking results and analytical use

HSAD is explicitly positioned as a stress test for pretrained audio models that appear strong on standard spoof benchmarks but fail under mixed-source conditions. In the HSAD analysis centered on transformer models, **pretrained models overgeneralize and collapse under hybrid conditions**, while **spoof-specific fine-tuning improves separability but struggles with unseen compositions**, and **dataset-specific adaptation on HSAD yields large performance gains** [2509.07323].

The strongest quantitative results reported for HSAD come from fine-tuned **Audio Spectrogram Transformer (AST)** models. In one study, **Model A**—a fine-tuned **MIT-AST**—reaches **97.88%** validation accuracy, while **Model B**—a fine-tuned **MattyB95** AST—reaches **98.08%** validation accuracy; both achieve **97%** test accuracy and **F1-score ≈ 99%** on the HSAD test split [2505.15136]. The same study reports that both models correctly identified **hybrid audio samples with 100% precision**, that **recall exceeds 88%** for **AI-generated and AI-cloned** classes, and that **human speech accuracy climbs to 84%** after HSAD fine-tuning [2505.15136]. It further reports a **71%** reduction in false positives for human speech for Model A, from **1,207 to 197**, and a **95%** reduction in false negatives for AI-generated speech for Model B, from **14,117 to 684** [2505.15136].

The later HSAD analysis broadens the model set to six transformer-based architectures, including **MIT-AST**, **MattyB95-AST**, **Wav2Vec2**, and **HuBERT** [2509.07323]. It reports a striking failure mode for pretrained **MIT-AST**: despite **93.67%** aggregate accuracy, the model **fails to correctly identify any genuine human utterances**, effectively treating everything as spoofed [2509.07323]. The same paper reports that pretrained **Wav2Vec2-base-960h**, **Wav2Vec2-large-960h**, and **HuBERT-base-ls960** achieve only **20.06%**, **3.52%**, and **1.30%** overall accuracy, respectively, before HSAD fine-tuning, and then rise to **97.94%**, **97.26%**, and **96.76%** after HSAD-specific adaptation [2509.07323]. The intended analytical conclusion is not merely that fine-tuning helps, but that HSAD reveals **calibration failures**, **class collapse**, and **bias toward global spoof cues** that standard binary datasets do not expose [2509.07323].

This empirical pattern closely aligns with the more general claim that models trained only on homogeneous real-versus-fake corpora tend to **overestimate** their robustness and **fail in hybrid scenarios**, especially when **only part of the signal is spoofed**, when spoofed segments are **short**, or when the utterance is **carefully blended** [2505.15136].

## 5. Relation to adjacent hybrid and partially spoofed corpora

HSAD belongs to a broader family of datasets that redefine spoofing as a mixed, structured, or partially localized phenomenon rather than a whole-utterance binary label.

| Resource | Hybrid notion | Key structure |
|---|---|---|
| PartialSpoof | bona fide and spoofed segments in one utterance | segment- and utterance-level labels |
| EchoFake | synthetic plus physical replay | B, RB, F, RF |
| CompSpoof | spoofed speech and/or spoofed environment | 5 component-level classes |
| LinguAS / 2024 hybrid dataset | multi-attack audio plus EDLFs | linguistic anomaly labels |
| Named HSAD | human, cloned, AI-generated, mixed audio | 4 classes with degraded variants |

**PartialSpoof** introduces a database in which utterances are only **partially spoofed**, with bona fide and spoofed segments mixed inside one file, and shows that countermeasures trained on fully spoofed utterances degrade substantially when tested on partially spoofed data, whereas training on partially spoofed data is reliable for both fully and partially spoofed utterances [2104.02518]. This is conceptually close to HSAD’s mixed human–synthetic utterances, although PartialSpoof is built from **ASVspoof 2019 LA** and is explicitly **segment-labeled** [2104.02518].

**EchoFake** extends the hybrid idea from within-utterance mixing to **attack-mode mixing**, combining **bona fide (B)**, **replayed bona fide (RB)**, **fake (F)**, and **replayed fake (RF)** speech under real acoustic playback and re-recording conditions [2510.19414]. Its role in the HSAD landscape is to show that hybridization can arise not only from segment concatenation but also from the interaction between synthesis and physical replay.

**CompSpoof** pushes the notion further toward **component-level audio spoofing**, where only the **speech** component, only the **environmental sound** component, both, or neither are spoofed [2509.15804]. Its five classes—**original**, **bonafide_bonafide**, **spoof_bonafide**, **bonafide_spoof**, and **spoof_spoof**—generalize the HSAD premise from segment mixtures to overlapping source components [2509.15804].

**LinguAS** and the earlier unnamed hybrid dataset demonstrate that hybrid spoof corpora can also be organized around **annotation richness** rather than only mixed-source synthesis. LinguAS contains **over 800 audio samples**, each annotated with five EDLFs and accompanied by **gender** and **generator metadata**, while the 2024 unnamed hybrid dataset contains **344 audio samples balanced in terms of spoofed vs genuine**, with **30% VC**, **30% TTS**, **25% replay attacks**, and **15% mimicry** [2606.10246][2409.06033]. In both cases, the hybrid property lies in combining multiple spoof families and adding a human-interpretable annotation layer.

At the design level, the major antecedent is **ASVspoof 2019**, which unifies **Logical Access (LA)** and **Physical Access (PA)** scenarios and treats **synthetic**, **converted**, and **replayed** speech within a common evaluation framework [1904.05441][1911.01601]. This multi-attack, known-versus-unknown philosophy has become a template for later hybrid spoof datasets, including HSAD-style resources.

## 6. Ambiguities, limitations, and future directions

The named HSAD literature is explicit about several limitations. The benchmark is built from only **12 speakers**, is **English** only, and derives its clean recordings from a **controlled indoor studio-like setup** rather than in-the-wild capture [2505.15136][2509.07323]. Its spoof coverage is centered on **TTS-based synthesis** and **AI voice cloning**, with limited or no explicit inclusion of **voice conversion**, **GAN-based attacks**, **prosody-only manipulations**, or **adversarial examples** [2505.15136][2509.07323]. The current evaluation remains largely **utterance-level**, even though hybrid constructions motivate finer-grained temporal localization [2505.15136][2509.07323].

A second limitation is terminological and documentary. One line of work describes a hybrid spoof corpus with rich sociolinguistic annotations but **does not** use the HSAD acronym at all [2409.06033]. Conversely, the 2025 HSAD literature uses the acronym explicitly but contains a visible discrepancy between an abstract claiming **41,044 degraded utterances** and a detailed construction section that consistently describes **1,248 degraded utterances** [2509.07323]. The existence of both statements in the same paper means that any encyclopedic treatment of HSAD must distinguish the dataset’s intended design from its unresolved reporting inconsistency.

The research directions proposed around HSAD are comparatively consistent. They include **expanding spoof modalities** to **voice conversion**, **more TTS engines**, **multilingual speech**, **GAN-based synthesis**, and **adversarially crafted attacks**; adding **real-world recordings** from **smartphones**, **smart speakers**, **telephony systems**, and varied acoustic environments; releasing **segment-level annotations** for localization; and conducting **cross-corpus benchmarking** against resources such as **HAD** and **ADD2023-PF** [2505.15136][2509.07323]. A related line of work on codec-resynthesized speech argues that future hybrid spoof datasets should also separate **clean human speech**, **benign codec/vocoder processing**, and **fully synthetic codec-based deepfakes**, because different labeling choices materially change detector behavior [2602.16343]. This suggests that the long-term evolution of HSAD is likely to involve not only larger scale and more attack families, but also more granular label taxonomies for ambiguous or borderline cases.

Within anti-spoofing research, HSAD is therefore best viewed as both a named benchmark and a broader design philosophy: a move away from homogeneous real-versus-fake corpora toward datasets that encode **mixed provenance**, **partial manipulation**, **cross-condition degradation**, and, increasingly, **interpretable metadata** that expose what current detectors actually learn [2505.15136][2509.07323][2409.06033].

Source: https://www.emergentmind.com/topics/hybrid-spoofed-audio-dataset-hsad