emotional-speech 开源项目推荐（19 个）

Plachtaa /VALL-E-X

Microsoft VALL-E X 零样本 TTS 模型的开源实现。演示可在 https://plachtaa.github.io/vallex/ 获取

abikaki /awesome-speech-emotion-recognition

😎 Awesome lists about Speech Emotion Recognition

RaduBolbo /F5-TTS-Emotional-CFG

Zero-shot voice cloning text-to-speech (TTS) with explicit emotion class conditioning built on F5-TTS

aaivu /EmoTa

EmoTa is an open-access Tamil Speech Emotion Recognition dataset with 936 utterances from 22 native speakers, covering five emotions (anger, happiness, sadness, fear, and neutrality). It supports emotion classification tasks and advances Tamil language processing.

aliyzd95 /modified-shemo

A modification on the Sharif Emotional Speech Database

ZET-Speech /ZET-Speech-Demo

ZET-Speech: Zero-shot adaptive Emotion-controllable Text-to-Speech Synthesis with Diffusion and Style-based Models (TTS)

Wonbin-Jung /e3-vits

Official GitHub page of E3-VITS

nipponjo /tts-german-pytorch

🎙️ German TTS (FastPitch) with Thorsten voice / emotional

SMIL-SPCRAS /EMOLIPS

EMOLIPS: TWO-LEVEL APPROACH FOR LIP-READING EMOTIONAL SPEECH

aaivu /KuralHub

Speech emotion recognition (SER) datasets for 70 languages and a benchmark of 11 speech models (HuBERT, wav2vec 2.0, WavLM, Whisper) in 29 languages. Interspeech 2026.

NewJerseyStyle /anime-translator

Applying deep learning to translate animation and re-generate audio.

lryanner /aichat

A GUI program for chat with chatbot such as chatgpt.

Conscht /TargetSEC

TargetSEC: Arousal guided style conversion using in-the-wild-data.

Mormolykos /bedvibe-datasets

Multilingual emotional speech datasets for TTS training

sirrauf /Edison-AT-AI-EmotionalDepression

Edison AT is AI emotional depression program. Developed using Python.

Shadowy-lionhunter5498 /transformer-architecture

Explore the Transformer and DeepSeek V4.1 Flash in interactive 3D, with guided tours and component-level breakdowns.

Mormolykos /speaker-encoders-disagree

Fourteen speaker-verification encoders on identical trials under expressive speech: EER spans 0.047-0.233 on the same audio. Score tables, analysis code, and ten retracted claims. Paper: 10.5281/zenodo.22158030

lou-ukw /sprechstil_haubenreiser

This is a project dedicated to the classification of emotional speech and was created in class with Prof. Dr. Burkhardt at Technische Universität Berlin.

lou-ukw /indivsprech_haubenreiser

This is a project dedicated to the classification of emotional speech and was created in class with Prof. Dr. Burkhardt at Technische Universität Berlin.