Electrical Engineering and Systems Science > Audio and Speech Processing
[Submitted on 30 Aug 2021 (v1), last revised 16 Feb 2022 (this version, v6)]
Title:Neural HMMs are all you need (for high-quality attention-free TTS)
View PDFAbstract:Neural sequence-to-sequence TTS has achieved significantly better output quality than statistical speech synthesis using HMMs. However, neural TTS is generally not probabilistic and uses non-monotonic attention. Attention failures increase training time and can make synthesis babble incoherently. This paper describes how the old and new paradigms can be combined to obtain the advantages of both worlds, by replacing attention in neural TTS with an autoregressive left-right no-skip hidden Markov model defined by a neural network. Based on this proposal, we modify Tacotron 2 to obtain an HMM-based neural TTS model with monotonic alignment, trained to maximise the full sequence likelihood without approximation. We also describe how to combine ideas from classical and contemporary TTS for best results. The resulting example system is smaller and simpler than Tacotron 2, and learns to speak with fewer iterations and less data, whilst achieving comparable naturalness prior to the post-net. Our approach also allows easy control over speaking rate.
Submission history
From: Gustav Eje Henter [view email][v1] Mon, 30 Aug 2021 15:38:00 UTC (438 KB)
[v2] Fri, 3 Sep 2021 16:56:00 UTC (439 KB)
[v3] Sat, 9 Oct 2021 10:39:25 UTC (450 KB)
[v4] Sun, 28 Nov 2021 11:13:20 UTC (449 KB)
[v5] Mon, 10 Jan 2022 22:01:46 UTC (450 KB)
[v6] Wed, 16 Feb 2022 16:56:08 UTC (373 KB)
Current browse context:
eess.AS
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.