qin/Paper_Read
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

README

Browse Source
This commit is contained in:
superlishunqin 2025-06-27 03:50:31 +08:00
parent 848c447125
commit 813ffc2aff
7 changed files with 8 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

0
2106.15561v3.pdf → A_Survey_on_Neural_Speech_Synthesis.pdf
View File

0
pone.0283440.pdf → A_real-time_voice_cloning_system_with_multiple_algorithms_for_speech_quality_improvement.pdf
View File

0
s13636-024-00329-7.pdf → Deep_learning-based_expressive_speech_synthesis-a_systematic_review_of_approaches_challenges_and_resources.pdf
View File

0
2006.04558v8.pdf → FASTSPEECH_2-FAST_AND_HIGH-QUALITY_END-TO-END_TEXT_TO_SPEECH.pdf
View File

0
2205.04421v2.pdf → NaturalSpeech-End-to-End_Text_to_Speech_Synthesis_with_Human-Level_Quality.pdf
View File

15
README.md
View File

@ -20,13 +20,13 @@
| 文件名 | 核心主题 | 阅读笔记 / 目的 | 论文标题 (参考) | 状态 | 链接 |
| :--- | :--- | :--- | :--- | :--- | :--- |
| `2106.15561v3.pdf` | TTS技术综述 | 此综述详细介绍了基于神经网络的TTS技术，用于构建对现代语音合成技术全景的理解。 | *A Survey on Neural Speech Synthesis* | 已归档 | [原文链接](https://arxiv.org/abs/2106.15561) |
| `s13636-024-00329-7.pdf` | 情感语音合成 | 此论文系统回顾了情感语音合成的方法、挑战和资源，与论文“融合情感”部分高度相关。 | *Deep learning-based expressive speech synthesis: a systematic review...* | 已归档 | [原文链接](https://asmp-eurasipjournals.springeropen.com/articles/10.1186/s13636-024-00329-7) |
| `A_Survey_on_Neural_Speech_Synthesis.pdf` | TTS技术综述 | 此综述详细介绍了基于神经网络的TTS技术，用于构建对现代语音合成技术全景的理解。 | *A Survey on Neural Speech Synthesis* | 已归档 | [原文链接](https://arxiv.org/abs/2106.15561) |
| `Deep_learning-based_expressive_speech_synthesis-a_systematic_review_of_approaches_challenges_and_resources.pdf` | 情感语音合成 | 此论文系统回顾了情感语音合成的方法、挑战和资源，与论文“融合情感”部分高度相关。 | *Deep learning-based expressive speech synthesis: a systematic review...* | 已归档 | [原文链接](https://asmp-eurasipjournals.springeropen.com/articles/10.1186/s13636-024-00329-7) |
| `Text_to_Speech_Synthesis_A_Systematic_Review_Deep_.pdf` | TTS架构与方向 | 此综述覆盖了深度学习TTS架构和未来研究方向，为技术选型和未来展望提供参考。 | *Text to Speech Synthesis: A Systematic Review, Deep Learning Based Architecture and Future Research Direction* | 已归档 | [原文链接](https://www.researchgate.net/publication/364280141_Text_to_Speech_Synthesis_A_Systematic_Review_Deep_Learning_Based_Architecture_and_Future_Research_Direction) |
| `2006.04558v8.pdf` | 非自回归TTS模型 | 用于理解以FastSpeech为代表的非自回归模型如何解决“一对多”问题，以及如何引入韵律等变化信息来提升合成质量与速度。 | *FASTSPEECH 2: FAST AND HIGH-QUALITY END-TO-END TEXT TO SPEECH* | 已归档 | [原文链接](https://paperswithcode.com/paper/fastspeech-2-fast-and-high-quality-end-to-end) |
| `2205.04421v2.pdf` | SOTA TTS模型 | 了解如何通过设计更强大的模型（如NaturalSpeech）和利用大规模语料库来实现与人类相媲美的合成效果，这能帮助理解当前技术的天花板在哪里。| *NaturalSpeech: End-to-End Text to Speech Synthesis with Human-Level Quality* | 已归档 | [原文链接](https://arxiv.org/abs/2205.04421) |
| `1806.04558v4.pdf` | 零样本/迁移学习TTS | 此论文讲述利用迁移学习技术，构建一个能够生成任意说话人（包括训练中未见过的说话人）声音的文本到语音（TTS）系统。 | *Transfer Learning from Speaker Verification to Multispeaker Text-To-Speech Synthesis* | 已归档 | [原文链接](https://paperswithcode.com/paper/transfer-learning-from-speaker-verification) |
| `pone.0283440.pdf` | 实时语音克隆 | 此综述旨在提升语音克隆质量的实时系统。 | *A real-time voice cloning system with multiple algorithms for speech quality improvement* | 已归档 | [原文链接](https://pmc.ncbi.nlm.nih.gov/articles/PMC10069766/) |
| `FASTSPEECH_2-FAST_AND_HIGH-QUALITY_END-TO-END_TEXT_TO_SPEECH.pdf` | 非自回归TTS模型 | 用于理解以FastSpeech为代表的非自回归模型如何解决“一对多”问题，以及如何引入韵律等变化信息来提升合成质量与速度。 | *FASTSPEECH 2: FAST AND HIGH-QUALITY END-TO-END TEXT TO SPEECH* | 已归档 | [原文链接](https://paperswithcode.com/paper/fastspeech-2-fast-and-high-quality-end-to-end) |
| `NaturalSpeech-End-to-End_Text_to_Speech_Synthesis_with_Human-Level_Quality.pdf` | SOTA TTS模型 | 了解如何通过设计更强大的模型（如NaturalSpeech）和利用大规模语料库来实现与人类相媲美的合成效果，这能帮助理解当前技术的天花板在哪里。| *NaturalSpeech: End-to-End Text to Speech Synthesis with Human-Level Quality* | 已归档 | [原文链接](https://arxiv.org/abs/2205.04421) |
| `Transfer_Learning_from_Speaker_Verification_to_Multispeaker_Text-To-Speech_Synthesis.pdf` | 零样本/迁移学习TTS | 此论文讲述利用迁移学习技术，构建一个能够生成任意说话人（包括训练中未见过的说话人）声音的文本到语音（TTS）系统。 | *Transfer Learning from Speaker Verification to Multispeaker Text-To-Speech Synthesis* | 已归档 | [原文链接](https://paperswithcode.com/paper/transfer-learning-from-speaker-verification) |
| `A_real-time_voice_cloning_system_with_multiple_algorithms_for_speech_quality_improvement.pdf` | 实时语音克隆 | 此综述旨在提升语音克隆质量的实时系统。 | *A real-time voice cloning system with multiple algorithms for speech quality improvement* | 已归档 | [原文链接](https://pmc.ncbi.nlm.nih.gov/articles/PMC10069766/) |
| `OpenVoice Versatile Instant Voice Cloning.pdf` | 情感可控的语音克隆 | 研究OpenVoice如何解耦音色与情感等风格，以实现对克隆声音的灵活情感控制，这与论文核心“融合情感的语音克隆”高度相关。 | *OpenVoice: Versatile Instant Voice Cloning* | 已归档 | [原文链接](https://arxiv.org/abs/2312.01479) |
| `Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech.pdf` | 端到端TTS模型 (VITS) | 理解VITS模型如何结合VAE和对抗学习，实现高质量的并行端到端语音合成，为技术选型提供重要参考。 | *Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech* | 已归档 | [原文链接](https://arxiv.org/abs/2106.06103) |
| `Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions.pdf` | 里程碑式TTS模型 (Tacotron 2) | 学习里程碑模型Tacotron 2的两阶段架构，理解其如何奠定高质量端到端语音合成的基础，为本研究提供技术背景和起点。 | *Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions* | 已归档 | [原文链接](https://arxiv.org/abs/1712.05884) |
@ -63,4 +63,5 @@
* 持续跟进SOTA（State-of-the-Art）语音克隆与情感合成模型，特别是低资源和高表现力的相关技术。
* 深入阅读儿童发展心理学、人机交互（HCI）中关于儿童与技术互动的相关文献。
* 关注大语言模型（LLM）在教育领域，特别是对话系统和个性化辅导方面的最新研究。
* 关注大语言模型（LLM）在教育领域，特别是对话系统和个性化辅导方面的最新研究。

0
1806.04558v4.pdf → Transfer_Learning_from_Speaker_Verification_to_Multispeaker_Text-To-Speech_Synthesis.pdf
View File