strob/gentle

gentle forced aligner

What it solves

It provides a way to align speech audio with a provided text transcript, allowing you to determine exactly when specific words are spoken in an audio file.

How it works

Built on the Kaldi speech recognition toolkit, Gentle acts as a forced-aligner. It takes an audio file and a corresponding text transcript as input and matches the sounds in the audio to the words in the text.

Who it’s for

Developers and users who need precise timing for spoken words in audio files, whether through a graphical user interface, a REST API for integration into other software, or a command-line tool.

Highlights

  • Robust and lenient forced-alignment of speech and text.
  • Multiple deployment options including a Mac app, Docker image, and source installation.
  • Accessible via a web GUI, REST API, or CLI.
  • Outputs alignment data in JSON format.

Related

  • Project
    MontrealCorpusTools/Montreal-Forced-Aligner

    A command-line utility for performing forced alignment of speech datasets, synchronizing audio recordings with their text transcripts using Kaldi.

  • Project
    MahmoudAshraf97/ctc-forced-aligner

    A Python package for efficient forced alignment of audio and text using Hugging Face CTC models, supporting over 1,100 languages.

  • Project
    readbeyond/aeneas

    A Python/C library and toolset for forced alignment, automatically synchronizing audio narrations with their corresponding text fragments.

  • Project
    daanzu/kaldi-active-grammar

    Kaldi‑Active‑Grammar is a Python wrapper for the Kaldi speech‑recognition engine that lets you compile many small grammars and activate only the needed ones per utterance. It ships binary wheels for all major OSes, includes a pre‑trained English model, and provides a backend for the Dragonfly and Caster voice‑control frameworks. Install via pip, download the model, define rules, and dynamically enable them for context‑aware command‑and‑control.

  • Project
    linto-ai/whisper-timestamped

    An extension of openai-whisper that provides accurate word-level timestamps and confidence scores for multilingual automatic speech recognition.