Methods › Audio › Text-to-Speech Models › FastPitch

FastPitch

9 papers tagged archive 2025-07-28

Introduced by Adrian Łańcucki in FastPitch: Parallel Text-to-speech with Pitch Prediction

archive 2025-07-28 Description, source and code snippet are the archive's method entry.

FastPitch is a fully-parallel text-to-speech model based on FastSpeech, conditioned on fundamental frequency contours. The architecture of FastPitch is shown in the Figure. It is based on FastSpeech and composed mainly of two feed-forward Transformer (FFTr) stacks. The first one operates in the resolution of input tokens, the second one in the resolution of the output frames. Let x=(x₁, …, xₙ) be the sequence of input lexical units, and 𝐲=(y₁, …, yₜ) be the sequence of target mel-scale spectrogram frames. The first FFTr stack produces the hidden representation 𝐡=FFTr(𝐱). The hidden representation h is used to make predictions about the duration and average pitch of every character with a 1-D CNN

𝐝̂= DurationPredictor (𝐡), 𝐩̂=PitchPredictor(𝐡)

where 𝐝̂ ∈ℕⁿ and 𝐩̂ ∈ℝⁿ. Next, the pitch is projected to match the dimensionality of the hidden representation h ∈ ℝ^(n ×d) and added to 𝐡. The resulting sum 𝐠 is discretely upsampled and passed to the output FFTr, which produces the output mel-spectrogram sequence

𝐠=𝐡+PitchEmbedding(𝐩)

𝐲̂=FFTr([g₁, …, g₁_(d₁), …gₙ, …, gₙ_(dₙ)])

Ground truth 𝐩 and 𝐝 are used during training, and predicted 𝐩̂ and 𝐝̂ are used during inference. The model optimizes mean-squared error (MSE) between the predicted and ground-truth modalities

ℒ=𝐲̂-𝐲₂²+α𝐩̂-𝐩₂²+γ𝐝̂-𝐝₂²

PaperSource

Papers archive 2025-07-28

9 shown of 9, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.

Tasks archive 2025-07-28

20 shown of 28 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.

TaskPapers
Text to Speech9
text-to-speech9
Speech Synthesis5
Text-To-Speech Synthesis4
Diversity2
Automatic Speech Recognition1
Automatic Speech Recognition (ASR)1
Disentanglement1
Machine Translation1
Prediction1
Sentence1
Speech Recognition1
Speech Synthesis - Assamese1
Speech Synthesis - Bengali1
Speech Synthesis - Bodo1
Speech Synthesis - Gujarati1
Speech Synthesis - Hindi1
Speech Synthesis - Kannada1
Speech Synthesis - Malayalam1
Speech Synthesis - Manipuri1

Usage over time archive 2025-07-28

Papers per year tagged with FastPitch: 2020 to 2024, peak 3 3 0 2020: 1 paper 2020 2021: 1 paper 2021 2022: 3 papers 2022 2023: 1 paper 2023 2024: 3 papers 2024
Papers per year the archive tags with this method, by the paper's archive date (9 dated). Bars are counts, not a trend claim.

Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).

Categories archive 2025-07-28

Text-to-Speech Models

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections