SOTAVerified

Speech Synthesis

Speech synthesis is the task of generating speech from some other modality like text, lip movements etc.

Please note that the leaderboards here are not really comparable between studies - as they use mean opinion score as a metric and collect different samples from Amazon Mechnical Turk.

( Image credit: WaveNet: A generative model for raw audio )

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 301–350 of 1249 papers

TitleStatusHype
Balancing Speech Understanding and Generation Using Continual Pre-training for Codec-based Speech LLM—0
High-Fidelity Music Vocoder using Neural Audio Codecs—0
AV-Flow: Transforming Text to Audio-Visual Human-like Interactions—0
A Survey on Bridging EEG Signals and Generative AI: From Image and Text to Beyond—0
NaturalL2S: End-to-End High-quality Multispeaker Lip-to-Speech Synthesis with Differential Digital Signal Processing—0
FELLE: Autoregressive Speech Synthesis with Token-Wise Coarse-to-Fine Flow Matching—0
ASVspoof 5: Design, Collection and Validation of Resources for Spoofing, Deepfake, and Adversarial Attack Detection Using Crowdsourced Speech—0
LoRP-TTS: Low-Rank Personalized Text-To-Speech—0
Non-invasive electromyographic speech neuroprosthesis: a geometric perspective—0
Gender Bias in Instruction-Guided Speech Synthesis Models—0
Continuous Autoregressive Modeling with Stochastic Monotonic Alignment for Speech Synthesis—0
Compact Neural TTS Voices for Accessibility—0
Generalizable Audio Deepfake Detection via Latent Space Refinement and Augmentation—0
Generative Data Augmentation Challenge: Zero-Shot Speech Synthesis for Personalized Speech Enhancement—0
Speech Synthesis along Perceptual Voice Quality Dimensions—0
A Non-autoregressive Model for Joint STT and TTS—0
Exploring the encoding of linguistic representations in the Fully-Connected Layer of generative CNNs for Speech—0
Low-Resource Text-to-Speech Synthesis Using Noise-Augmented Training of ForwardTacotron—0
TTS-Transducer: End-to-End Speech Synthesis with Neural Transducer—0
PROEMO: Prompt-Driven Text-to-Speech Synthesis Based on Emotion and Intensity Control—0
Probing Speaker-specific Features in Speaker Representations—0
JELLY: Joint Emotion Recognition and Context Reasoning with LLMs for Conversational Speech Synthesis—0
FleSpeech: Flexibly Controllable Speech Generation with Various Prompts—0
FaceSpeak: Expressive and High-Quality Speech Synthesis from Human Portraits of Different Styles—0
Stable-TTS: Stable Speaker-Adaptive Text-to-Speech Synthesis via Prosody Prompting—0
CrossSpeech++: Cross-lingual Speech Synthesis with Decoupled Language and Speaker Generation—0
VoiceDiT: Dual-Condition Diffusion Transformer for Environment-Aware Speech Synthesis—0
MRI2Speech: Speech Synthesis from Articulatory Movements Recorded by Real-time MRI—0
Intra- and Inter-modal Context Interaction Modeling for Conversational Speech SynthesisCode0
Incremental Disentanglement for Environment-Aware Zero-Shot Text-to-Speech Synthesis—0
Autoregressive Speech Synthesis with Next-Distribution Prediction—0
Deep Speech Synthesis from Multimodal Articulatory Representations—0
ProsodyFM: Unsupervised Phrasing and Intonation Control for Intelligible Speech Synthesis—0
Efficient Generative Modeling with Residual Vector Quantization-Based Tokens—0
AMuSeD: An Attentive Deep Neural Network for Multimodal Sarcasm Detection Incorporating Bi-modal Data Augmentation—0
Multimodal Latent Language Modeling with Next-Token DiffusionCode0
Zero-Shot Mono-to-Binaural Speech Synthesis—0
Analytic Study of Text-Free Speech Synthesis for Raw Audio using a Self-Supervised Learning Model—0
Visatronic: A Multimodal Decoder-Only Model for Speech Synthesis—0
VQalAttent: a Transparent Speech Generation Pipeline based on Transformer-learned VQ-VAE Latent Space—0
Debatts: Zero-Shot Debating Text-to-Speech Synthesis—0
Complete reconstruction of the tongue contour through acoustic to articulatory inversion using real-time MRI data—0
Augmenting Polish Automatic Speech Recognition System With Synthetic Data—0
Fast and High-Quality Auto-Regressive Speech Synthesis via Speculative Decoding—0
Get Large Language Models Ready to Speak: A Late-fusion Approach for Speech Generation—0
Making Social Platforms Accessible: Emotion-Aware Speech Generation with Integrated Text Analysis—0
Continuous Speech Synthesis using per-token Latent Diffusion—0
A Unified Framework for Collecting Text-to-Speech Synthesis Datasets for 22 Indian Languages—0
Accelerating Codec-based Speech Synthesis with Multi-Token Prediction and Speculative Decoding—0
DurIAN-E 2: Duration Informed Attention Network with Adaptive Variational Autoencoder and Adversarial Learning for Expressive Text-to-Speech Synthesis—0
Show:102550
← PrevPage 7 of 25Next →
All datasetsLibriTTSNorth American EnglishLJSpeechMandarin ChineseBlizzard Challenge 2013

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1PeriodWave-Turbo-LPESQ4.45—Unverified
2BigVGAN-v2PESQ4.36—Unverified
3EVA-GAN-bigPESQ4.35—Unverified
4PeriodWave + FreeUPESQ4.25—Unverified
5RFWavePESQ4.23—Unverified
6BigVSAN (w/ snakebeta)PESQ4.12—Unverified
7BigVSANPESQ4.12—Unverified
8EVA-GAN-basePESQ4.03—Unverified
9BigVGANPESQ4.03—Unverified
10VocosPESQ3.7—Unverified
#ModelMetricClaimedVerifiedStatus
1Tacotron 2Mean Opinion Score4.53—Unverified
2WaveNet (Linguistic)Mean Opinion Score4.34—Unverified
3WaveNet (L+F)Mean Opinion Score4.21—Unverified
4TacotronMean Opinion Score4—Unverified
5HMM-driven concatenativeMean Opinion Score3.86—Unverified
6LSTM-RNN parametricMean Opinion Score3.67—Unverified
7meansMean Opinion Score0—Unverified
#ModelMetricClaimedVerifiedStatus
1BDDM vocoderMean Opinion Score4.48—Unverified
2DiffWave LARGEMean Opinion Score4.44—Unverified
3Neural HMMMean Opinion Score3.24—Unverified
4Neural HMM Ablation with 1 state per phoneMean Opinion Score2.68—Unverified
#ModelMetricClaimedVerifiedStatus
1WaveNet (L+F)Mean Opinion Score4.08—Unverified
2LSTM-RNN parametricMean Opinion Score3.79—Unverified
3HMM-driven concatenativeMean Opinion Score3.47—Unverified
#ModelMetricClaimedVerifiedStatus
1SampleRNN (2-tier)NLL1.39—Unverified
2SampleRNN (3-tier)NLL1.39—Unverified