Audio & Voice

Generate expressive speech and clone an authorized voice locally with an efficient open-source TTS model.

Turn lyrics and musical direction into a complete song with an open full-song generation model.

A local toolkit for text-to-music, music continuation, song, and audio generation with long-form output support.

Turn a lyric or music idea into a quick generated song sketch.

Turn timestamped lyrics and a style prompt into a full song with vocals and accompaniment.

Generate and edit songs from lyrics, tags, and audio ideas with an open music model.

Translate speech as it is spoken while keeping a natural streaming rhythm and optional voice characteristics.

Turn speech into text with a fast open transcription model.

Run multilingual speech recognition or translation with an open NVIDIA model.

Generate sound that follows the action in a video.

Transcribe speech and get timing for every word, not just every segment.

Add and detect localized watermarks in AI-generated speech with an open-source Python toolkit.

Turn a recording into text for captions, notes, and content repurposing in a few minutes.

Turn a video, social link, or audio file into a transcript, subtitles, and a summary.

Trim and export an audio clip in the browser without installing an editor.

Listen to a document or webpage instead of reading it on screen.

Turn a music library into searchable stems, loops, tempo, key, and MIDI data.