mlsanityData to 19 Oct 2026

Probing Human Articulatory Constraints in End-to-End TTS with Reverse and Mismatched Speech-Text Directions

arXiv 2602.14664 ↗First posted 16 Feb 2026cs.SDcited by 0

On the map: Speech and audio language models → Vocoders and speech synthesis

Abstract (arXiv)

An end-to-end (e2e) text-to-speech (TTS) system is a deep architecture that learns to associate a text string with acoustic speech patterns from a curated dataset. It is expected that all aspects associated with speech production, such as phone duration, speaker characteristics, and intonation among other things are captured in the trained TTS model to enable the synthesized speech to be natural and intelligible. Human speech is complex, involving smooth transitions between articulatory configurations (ACs). Due to anatomical constraints, some ACs are challenging to mimic or transition between. In this paper, we experimentally study if the constraints imposed by human anatomy have an implication on training an e2e-TTS systems. We experiment with two e2e-TTS architectures, namely, Tacotron-2 an autoregressive model and VITS-TTS a non-autoregressive model. In this study, we build TTS systems using (a) forward text, forward speech (conventional, e2e-TTS), (b) reverse text, reverse speech (r-e2e-TTS), and (c) reverse text, forward speech (rtfs-e2e-TTS). Experiments demonstrate that e2e-TTS systems are purely data-driven. Interestingly, the generated speech by r-e2e-TTS systems exhibits better fidelity, better perceptual intelligibility, and better naturalness

Key facts. No citing papers in the map yet (data through 19 Oct 2026). Placed in Vocoders and speech synthesis, one of 11 directions in Speech and audio language models. Builds on 11 of its 21 references that are in the map. First posted on arXiv 16 Feb 2026.

Builds on

11 of 21 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching9 Oct
  2. 2021Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech11 Jun

Other references · 9

  1. 2022Robust Speech Recognition via Large-Scale Weak Supervisioncited by 4,661
  2. 2014Neural Machine Translation by Jointly Learning to Align and Translatecited by 4,659
  3. 2015Neural Machine Translation of Rare Words with Subword Unitscited by 2,995
  4. 2018SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processingcited by 2,060
  5. 2017Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictionscited by 1,023
  6. 2018WaveGlow: A Flow-based Generative Network for Speech Synthesiscited by 365
  7. 2020Glow-TTS: A Generative Flow for Text-to-Speech via Monotonic Alignment Searchcited by 360
  8. 2023WhisperX: Time-Accurate Speech Transcription of Long-Form Audiocited by 359
  9. 2021Automatic Speaker Independent Dysarthric Speech Intelligibility Assessment Systemcited by 5

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Articulatory Source-Filter TTS: Physically Grounded Control through Vocal Tract Kinematics
  2. 2021Hierarchical prosody modeling and control in non-autoregressive parallel neural TTS
  3. 2021Non-Autoregressive TTS with Explicit Duration Modelling for Low-Resource Highly Expressive Speech
  4. 2022Deep Speech Synthesis from Articulatory Representations
  5. 2023E3 TTS: Easy End-to-End Diffusion-based Text to Speech
  6. 2026Training a Compact, Strong, and Realistic Text-to-Speech Model
  7. 2022Prosody-controllable spontaneous TTS with neural HMMs
  8. 2021Prosody-TTS: An end-to-end speech synthesis system with prosody control