Back: Catalog

Amphion

Amphion is an open-source toolkit developed for audio, music, and speech production. It aims to support reproducible research and present architectural visualizations of classical models to startup researchers. It supports many production tasks, from text to sound, song sound synthesis and sound transformation.

AM

Amphion

54

open-mmlab · Voice & Text-to-Speech

STARS10.3K
FORKS855
COMMITS0
CONTRIBUTORS0
VERSIONv0.1.1-alpha
UPDATED7 mo ago
PythonAudio GenerationTTSVocoderMIT

Common commercial products in this category: ElevenLabs, Amazon Polly, Google Cloud Text-to-Speech

What can it do?
  • You can perform text-based voice or speech synthesis tasks.
  • You can produce high-quality sound signals with different neural vokoder modules.
  • You can develop audio conversion and accent exchange applications.
Heads up
  • • Website is unreachable (checked: 2026-10-31)
User reviews
  • • NeutralGitHub Issue

    An issue with the preprocessing part of LibriTTS.

    “Hello!🤗 When I'm working with the LibriTTS dataset, I noticed that the generated test.json and train.json files do not include the corresponding text for each audio sample. This absence of text information causes an error when trying to extract the phonemes later on.…”

GitHub

My note

Alternative pages

ElevenLabs alternativesAmazon Polly alternativesGoogle Cloud Text-to-Speech alternatives

More in: Voice & Text-to-Speech