Skip to content

The truly free, open-source, local-first, cross-platform CapCut replacement. Scriptable through a JSON-RPC API & MCP.

feat(media): shelve read-aloud voices under Generated › Speech

Jareerd48de78b0d4c
diff --git a/docs/api/types.md b/docs/api/types.md
index e6d30960..a5631ad1 100644
--- a/docs/api/types.md
+++ b/docs/api/types.md
@@ -198,6 +198,7 @@ meaningfully even when the file is missing.
| `audioTracks` | *AudioTrack*[] | *optional*. Every audio stream, in file order |
| `placeholder` | bool | *optional*. True for a template slot |
| `colorRange` | string | *optional*. `limited` or `full`: the levels the picture is read as, over the file's own tag. Absent means "as tagged". Set with `setMediaColorRange` |
+| `origin` | string | *optional*. `speech` for a file the speech sheet read aloud. Absent for an import. The bin shelves a file with an origin under Generated, not with the imports |
### AudioTrack
@@ -304,7 +305,7 @@ A probed file, as `addMedia`, `fillSlot`, `replaceClipMedia` and
`freezeFrame` take it. The same fields as *MediaItem* minus `id` and
`placeholder`: `path`, `name`, `duration`, `kind`, `width`, `height`,
`frameRate`, `frameRateFraction`, `videoCodec`, `audioCodec`, `hasAudio`,
-`audioTracks`.
+`audioTracks`, and `origin` (*optional*; leave it out for an import).
> [!TIP]
> Build one from a `media.probe` reply: copy `path`, `duration`, `kind`
diff --git a/src/crates/concat-api/src/lib.rs b/src/crates/concat-api/src/lib.rs
index 0ae510eb..d3b7f843 100644
--- a/src/crates/concat-api/src/lib.rs
+++ b/src/crates/concat-api/src/lib.rs
@@ -792,6 +792,7 @@ mod tests {
audio_codec: None,
has_audio: false,
audio_tracks: Vec::new(),
+ origin: None,
}
}
diff --git a/src/crates/concat-export/src/flatten.rs b/src/crates/concat-export/src/flatten.rs
index b8d86aa2..508053f2 100644
--- a/src/crates/concat-export/src/flatten.rs
+++ b/src/crates/concat-export/src/flatten.rs
@@ -261,6 +261,7 @@ mod tests {
audio_codec: None,
has_audio: true,
audio_tracks: Vec::new(),
+ origin: None,
},
})
.expect("adds media")
diff --git a/src/crates/concat-host/src/media.rs b/src/crates/concat-host/src/media.rs
index 75809e1f..18007a34 100644
--- a/src/crates/concat-host/src/media.rs
+++ b/src/crates/concat-host/src/media.rs
@@ -124,6 +124,7 @@ impl MediaSummary {
.iter()
.map(AudioStreamInfo::to_track)
.collect(),
+ origin: None,
}
}
}
@@ -1027,6 +1028,7 @@ mod tests {
audio_codec: None,
has_audio: false,
audio_tracks: Vec::new(),
+ origin: None,
},
})
.expect("adds")
diff --git a/src/crates/concat-host/src/templates.rs b/src/crates/concat-host/src/templates.rs
index d1ad8d30..04d6df45 100644
--- a/src/crates/concat-host/src/templates.rs
+++ b/src/crates/concat-host/src/templates.rs
@@ -444,6 +444,7 @@ mod tests {
audio_codec: None,
has_audio: false,
audio_tracks: Vec::new(),
+ origin: None,
},
}
}
diff --git a/src/crates/concat-perf/src/main.rs b/src/crates/concat-perf/src/main.rs
index 8d02454c..d4394482 100644
--- a/src/crates/concat-perf/src/main.rs
+++ b/src/crates/concat-perf/src/main.rs
@@ -165,6 +165,7 @@ fn ripple_delete_of_200() -> Measure {
audio_codec: None,
has_audio: false,
audio_tracks: Vec::new(),
+ origin: None,
},
})
.expect("adds media")
@@ -256,6 +257,7 @@ fn undo_200_edits() -> Measure {
audio_codec: None,
has_audio: false,
audio_tracks: Vec::new(),
+ origin: None,
},
})
.expect("adds media")
@@ -305,6 +307,7 @@ fn document_round_trip() -> Measure {
audio_codec: None,
has_audio: false,
audio_tracks: Vec::new(),
+ origin: None,
},
})
.expect("adds media")
diff --git a/src/crates/concat-project/src/commands/clips.rs b/src/crates/concat-project/src/commands/clips.rs
index 5051cd60..5c3d5cd1 100644
--- a/src/crates/concat-project/src/commands/clips.rs
+++ b/src/crates/concat-project/src/commands/clips.rs
@@ -343,6 +343,7 @@ pub(super) fn apply(
audio_codec: item.audio_codec,
has_audio: item.has_audio,
audio_tracks: item.audio_tracks,
+ origin: item.origin,
placeholder: false,
color_range: None,
extra: Default::default(),
@@ -423,6 +424,7 @@ pub(super) fn apply(
audio_codec: None,
has_audio: false,
audio_tracks: Vec::new(),
+ origin: item.origin,
placeholder: false,
color_range: None,
extra: Default::default(),
diff --git a/src/crates/concat-project/src/commands/media.rs b/src/crates/concat-project/src/commands/media.rs
index bc153387..d0c7348b 100644
--- a/src/crates/concat-project/src/commands/media.rs
+++ b/src/crates/concat-project/src/commands/media.rs
@@ -38,6 +38,7 @@ pub(super) fn apply(
audio_codec: item.audio_codec,
has_audio: item.has_audio,
audio_tracks: item.audio_tracks,
+ origin: item.origin,
placeholder: false,
color_range: None,
extra: Default::default(),
diff --git a/src/crates/concat-project/src/commands/mod.rs b/src/crates/concat-project/src/commands/mod.rs
index adf3a642..861cfb76 100644
--- a/src/crates/concat-project/src/commands/mod.rs
+++ b/src/crates/concat-project/src/commands/mod.rs
@@ -14,8 +14,8 @@ use serde::{Deserialize, Serialize};
use crate::model::{
AnimationSlot, AppliedFilter, AudioTrack, Clip, ClipAnimation, ClipKind, ColorRange, Crop,
- CustomFont, Cutout, CutoutMode, KeyEase, KeyProperty, MediaItem, MediaKind, Project,
- SpeedPoint, Stroke, TextStyle, Timeline, Track, Transition, VideoSettings,
+ CustomFont, Cutout, CutoutMode, KeyEase, KeyProperty, MediaItem, MediaKind, MediaOrigin,
+ Project, SpeedPoint, Stroke, TextStyle, Timeline, Track, Transition, VideoSettings,
};
mod audio;
@@ -208,6 +208,11 @@ pub struct NewMedia {
/// Defaulted so a caller from before the list can still add media.
#[serde(default)]
pub audio_tracks: Vec<AudioTrack>,
+ /// What made the file, when the editor did; see `MediaItem::origin`.
+ /// Defaulted so an import, and a caller from before origins, says
+ /// nothing.
+ #[serde(default)]
+ pub origin: Option<MediaOrigin>,
}
/// Every edit, as the window sends it: a tagged `op` plus camelCase
diff --git a/src/crates/concat-project/src/lib.rs b/src/crates/concat-project/src/lib.rs
index 0351db5c..44b9eda7 100644
--- a/src/crates/concat-project/src/lib.rs
+++ b/src/crates/concat-project/src/lib.rs
@@ -40,7 +40,9 @@ mod tests {
use crate::commands::{ClipMove, ClipPatch, Command, NewMedia, TrackFlag, TrimEdge};
use crate::doc::DocumentSettings;
use crate::editor::Editor;
- use crate::model::{AudioTrack, ClipKind, MediaItem, MediaKind, Project, TextStyle};
+ use crate::model::{
+ AudioTrack, ClipKind, MediaItem, MediaKind, MediaOrigin, Project, TextStyle,
+ };
fn media(path: &str, duration: f64, has_audio: bool) -> Command {
Command::AddMedia {
@@ -57,6 +59,7 @@ mod tests {
audio_codec: has_audio.then(|| "aac".to_owned()),
has_audio,
audio_tracks: Vec::new(),
+ origin: None,
},
}
}
@@ -633,6 +636,7 @@ mod tests {
audio_codec: Some("aac".into()),
has_audio: true,
audio_tracks: Vec::new(),
+ origin: None,
};
let outcome = editor
.apply(Command::ReplaceClipMedia {
@@ -696,6 +700,7 @@ mod tests {
audio_codec: None,
has_audio: false,
audio_tracks: Vec::new(),
+ origin: None,
}
};
assert!(
@@ -737,6 +742,7 @@ mod tests {
audio_codec: None,
has_audio: false,
audio_tracks: Vec::new(),
+ origin: None,
};
let freeze_id = editor
.apply(Command::FreezeFrame {
@@ -830,6 +836,7 @@ mod tests {
audio_codec: None,
has_audio: false,
audio_tracks: Vec::new(),
+ origin: None,
}),
})
.expect("freezes")
@@ -2347,6 +2354,7 @@ mod tests {
audio_codec: None,
has_audio: false,
audio_tracks: Vec::new(),
+ origin: None,
},
})
.expect("fills");
@@ -2386,6 +2394,7 @@ mod tests {
audio_codec: None,
has_audio: false,
audio_tracks: Vec::new(),
+ origin: None,
},
});
assert!(
@@ -2480,6 +2489,57 @@ mod tests {
assert!(!editor.project().media[0].placeholder);
}
+ #[test]
+ fn a_medias_origin_round_trips_and_an_unknown_one_reads_as_an_import() {
+ let mut editor = Editor::new();
+ let Command::AddMedia { mut item } = media("/voice.wav", 3.0, true) else {
+ unreachable!()
+ };
+ item.kind = MediaKind::Audio;
+ item.origin = Some(MediaOrigin::Speech);
+ editor.apply(Command::AddMedia { item }).expect("adds");
+ editor.apply(media("/a.mp4", 10.0, true)).expect("adds");
+
+ let document = editor.to_document(&settings());
+ let media = document["media"].as_array().expect("a list");
+ assert_eq!(media[0]["origin"], json!("speech"));
+ assert!(
+ media[1].get("origin").is_none(),
+ "an import says nothing, so a document without voices stays byte-identical"
+ );
+ let restored = Editor::from_document(&document).expect("loads");
+ assert_eq!(
+ restored.project().media[0].origin,
+ Some(MediaOrigin::Speech)
+ );
+ assert_eq!(restored.project().media[1].origin, None);
+
+ // A document from a build that predates origins, and one from a
+ // build with origins this one has not heard of: both load, and
+ // both files are imports here.
+ let legacy = json!({
+ "name": "Old", "version": 1,
+ "media": [
+ { "id": "m1", "path": "/a.mp4", "name": "a.mp4", "kind": "video",
+ "hasAudio": false },
+ { "id": "m2", "path": "/b.wav", "name": "b.wav", "kind": "audio",
+ "hasAudio": true, "origin": "telepathy" },
+ { "id": "m3", "path": "/c.wav", "name": "c.wav", "kind": "audio",
+ "hasAudio": true, "origin": 7 }
+ ],
+ "tracks": [{ "id": "T1", "name": "Track 1", "visible": true, "muted": false }],
+ "clips": []
+ });
+ let editor = Editor::from_document(&legacy).expect("loads");
+ assert!(
+ editor
+ .project()
+ .media
+ .iter()
+ .all(|item| item.origin.is_none())
+ );
+ }
+
#[test]
fn a_garbage_document_is_none_not_a_panic() {
assert!(Editor::from_document(&json!("nonsense")).is_none());
@@ -3584,6 +3644,7 @@ mod tests {
audio_tracks: vec![],
placeholder: false,
color_range: None,
+ origin: None,
extra: Default::default(),
});
@@ -3621,6 +3682,7 @@ mod tests {
audio_tracks: vec![],
placeholder: false,
color_range: None,
+ origin: None,
extra: Default::default(),
});
diff --git a/src/crates/concat-project/src/model.rs b/src/crates/concat-project/src/model.rs
index 200edb1d..2dd5ba69 100644
--- a/src/crates/concat-project/src/model.rs
+++ b/src/crates/concat-project/src/model.rs
@@ -31,6 +31,16 @@ pub enum MediaKind {
Image,
}
+/// What made a media file, when it was the editor and not the user's
+/// import. The bin shelves such a file under Generated, by origin, and
+/// keeps it out of the import shelves.
+#[derive(Clone, Copy, PartialEq, Eq, Debug, Serialize, Deserialize)]
+#[serde(rename_all = "lowercase")]
+pub enum MediaOrigin {
+ /// Read aloud by the speech sheet.
+ Speech,
+}
+
/// What a clip can be - wider than [`MediaKind`] because a text clip has no
/// file behind it; it *is* its own content.
#[derive(Clone, Copy, PartialEq, Eq, Debug, Serialize, Deserialize)]
@@ -155,6 +165,14 @@ pub struct MediaItem {
#[serde(default, skip_serializing_if = "Option::is_none")]
#[serde(deserialize_with = "wire::maybe")]
pub color_range: Option<ColorRange>,
+ /// Where the file came from when the editor made it; see
+ /// [`MediaOrigin`]. Absent, and left out of the document, for an
+ /// import, and read as absent when it names an origin this build does
+ /// not know - the file is still a file, just one shelved with the
+ /// imports.
+ #[serde(default, skip_serializing_if = "Option::is_none")]
+ #[serde(deserialize_with = "wire::maybe")]
+ pub origin: Option<MediaOrigin>,
/// Fields this build does not know, kept so a document written by a
/// newer or a different build round-trips through this one intact.
#[serde(flatten, default, skip_serializing_if = "Map::is_empty")]
diff --git a/src/crates/concat/locales/de.json b/src/crates/concat/locales/de.json
index 73cd1fce..f12a1f38 100644
--- a/src/crates/concat/locales/de.json
+++ b/src/crates/concat/locales/de.json
@@ -235,6 +235,7 @@
"Go to end": "Zum Ende",
"Go to start": "Zum Anfang",
"Green": "Grün",
+ "Generated": "Generiert",
"Green Screen": "Greenscreen",
"Half": "Halb",
"Hall": "Saal",
@@ -340,6 +341,7 @@
"Nothing to read yet": "Noch nichts zu lesen",
"Nothing under the playhead": "Nichts unter dem Abspielkopf",
"Nothing was said in that clip": "In diesem Clip wurde nichts gesagt",
+ "Nothing read aloud yet. Text to speech puts its voice here.": "Noch nichts vorgelesen. Text zu Sprache legt seine Stimme hier ab.",
"Object": "Objekt",
"off": "aus",
"One repeat, fading.": "Eine Wiederholung, ausklingend.",
diff --git a/src/crates/concat/locales/en.json b/src/crates/concat/locales/en.json
index 5eb42380..e66d8d4f 100644
--- a/src/crates/concat/locales/en.json
+++ b/src/crates/concat/locales/en.json
@@ -369,6 +369,7 @@
"Generate Captions": "Generate Captions",
"Generate captions": "Generate captions",
"Generate speech": "Generate speech",
+ "Generated": "Generated",
"Generating voice": "Generating voice",
"Glitch": "Glitch",
"Glitch Warp": "Glitch Warp",
@@ -560,6 +561,7 @@
"not installed": "not installed",
"Nothing found to keep here, so the picture shows as shot. Try another subject, or custom removal.": "Nothing found to keep here, so the picture shows as shot. Try another subject, or custom removal.",
"Nothing of that kind in this project.": "Nothing of that kind in this project.",
+ "Nothing read aloud yet. Text to speech puts its voice here.": "Nothing read aloud yet. Text to speech puts its voice here.",
"Nothing starred yet — hover a tile and press its star.": "Nothing starred yet — hover a tile and press its star.",
"Nothing to caption yet": "Nothing to caption yet",
"Nothing to export": "Nothing to export",
diff --git a/src/crates/concat/locales/es.json b/src/crates/concat/locales/es.json
index 9ff38053..ec5c68fc 100644
--- a/src/crates/concat/locales/es.json
+++ b/src/crates/concat/locales/es.json
@@ -235,6 +235,7 @@
"Go to end": "Ir al final",
"Go to start": "Ir al inicio",
"Green": "Verde",
+ "Generated": "Generado",
"Green Screen": "Pantalla verde",
"Half": "Media",
"Hall": "Sala",
@@ -340,6 +341,7 @@
"Nothing to read yet": "Aún no hay nada que leer",
"Nothing under the playhead": "Nada bajo el cabezal",
"Nothing was said in that clip": "No se dijo nada en ese clip",
+ "Nothing read aloud yet. Text to speech puts its voice here.": "Aún no se ha leído nada. Texto a voz deja aquí su voz.",
"Object": "Objeto",
"off": "desactivado",
"One repeat, fading.": "Una repetición que se desvanece.",
diff --git a/src/crates/concat/locales/fa.json b/src/crates/concat/locales/fa.json
index d529a465..2cb9eddb 100644
--- a/src/crates/concat/locales/fa.json
+++ b/src/crates/concat/locales/fa.json
@@ -272,6 +272,7 @@
"Generate Captions": "ساخت زیرنویس‌ها",
"Generate captions": "ساخت زیرنویس‌ها",
"Generate speech": "ساخت گفتار",
+ "Generated": "تولیدشده",
"Generating voice": "در حال ساخت صدا",
"Glow": "درخشش",
"Go to end": "رفتن به انتها",
@@ -417,6 +418,7 @@
"not installed": "نصب نشده",
"Nothing found to keep here, so the picture shows as shot. Try another subject, or custom removal.": "چیزی برای نگه‌داشتن پیدا نشد، بنابراین تصویر به حالت اصلی نمایش داده می‌شود. سوژه دیگری را امتحان کنید یا از حذف سفارشی استفاده کنید.",
"Nothing of that kind in this project.": "هیچ موردی از این نوع در این پروژه وجود ندارد.",
+ "Nothing read aloud yet. Text to speech puts its voice here.": "هنوز چیزی خوانده نشده است. تبدیل متن به گفتار صدای خود را اینجا می‌گذارد.",
"Nothing starred yet — hover a tile and press its star.": "هنوز چیزی ستاره نشده است؛ نشانگر را روی کاشی ببرید و ستاره آن را بزنید.",
"Nothing to caption yet": "هنوز چیزی برای زیرنویس کردن وجود ندارد",
"Nothing to export": "چیزی برای خروجی گرفتن وجود ندارد",
diff --git a/src/crates/concat/locales/fr.json b/src/crates/concat/locales/fr.json
index 76336a76..607c9bc0 100644
--- a/src/crates/concat/locales/fr.json
+++ b/src/crates/concat/locales/fr.json
@@ -235,6 +235,7 @@
"Go to end": "Aller à la fin",
"Go to start": "Aller au début",
"Green": "Vert",
+ "Generated": "Généré",
"Green Screen": "Fond vert",
"Half": "Moitié",
"Hall": "Salle",
@@ -340,6 +341,7 @@
"Nothing to read yet": "Rien à lire pour l'instant",
"Nothing under the playhead": "Rien sous la tête de lecture",
"Nothing was said in that clip": "Rien n'a été dit dans ce clip",
+ "Nothing read aloud yet. Text to speech puts its voice here.": "Rien n'a encore été lu. La synthèse vocale dépose sa voix ici.",
"Object": "Objet",
"off": "désactivé",
"One repeat, fading.": "Une répétition qui s'estompe.",
diff --git a/src/crates/concat/locales/hr.json b/src/crates/concat/locales/hr.json
index 301f2def..28ad721b 100644
--- a/src/crates/concat/locales/hr.json
+++ b/src/crates/concat/locales/hr.json
@@ -286,6 +286,7 @@
"Go to end": "Idi na kraj",
"Go to start": "Idi na početak",
"Golden Hour": "Zlatni sat",
+ "Generated": "Generirano",
"Grain": "Zrno",
"Green": "Zelena",
"Green Screen": "Zelena pozadina",
@@ -424,6 +425,7 @@
"Nothing to export": "Ništa za izvoz",
"Nothing to read yet": "Još ništa za čitanje",
"Nothing under the playhead": "Ništa pod playheadom",
+ "Nothing read aloud yet. Text to speech puts its voice here.": "Još ništa nije pročitano naglas. Tekst u govor ovdje ostavlja svoj glas.",
"Nothing was said in that clip": "U tom klipu ništa nije rečeno",
"Object": "Objekt",
"off": "isključeno",
diff --git a/src/crates/concat/locales/it.json b/src/crates/concat/locales/it.json
index a9f478f4..23059204 100644
--- a/src/crates/concat/locales/it.json
+++ b/src/crates/concat/locales/it.json
@@ -235,6 +235,7 @@
"Go to end": "Vai alla fine",
"Go to start": "Vai all'inizio",
"Green": "Verde",
+ "Generated": "Generato",
"Green Screen": "Green screen",
"Half": "Metà",
"Hall": "Sala",
@@ -340,6 +341,7 @@
"Nothing to read yet": "Ancora niente da leggere",
"Nothing under the playhead": "Niente sotto la testina",
"Nothing was said in that clip": "In quella clip non è stato detto nulla",
+ "Nothing read aloud yet. Text to speech puts its voice here.": "Non è stato ancora letto nulla. La sintesi vocale lascia qui la sua voce.",
"Object": "Oggetto",
"off": "disattivato",
"One repeat, fading.": "Una ripetizione che sfuma.",
diff --git a/src/crates/concat/locales/ja.json b/src/crates/concat/locales/ja.json
index b555f6fc..a3c60858 100644
--- a/src/crates/concat/locales/ja.json
+++ b/src/crates/concat/locales/ja.json
@@ -235,6 +235,7 @@
"Go to end": "終端へ",
"Go to start": "先頭へ",
"Green": "緑",
+ "Generated": "生成",
"Green Screen": "グリーンスクリーン",
"Half": "1/2",
"Hall": "ホール",
@@ -340,6 +341,7 @@
"Nothing to read yet": "読み上げる内容がまだありません",
"Nothing under the playhead": "再生ヘッドの下に何もありません",
"Nothing was said in that clip": "そのクリップには発話がありませんでした",
+ "Nothing read aloud yet. Text to speech puts its voice here.": "まだ何も読み上げていません。テキスト読み上げの音声はここに入ります。",
"Object": "物体",
"off": "オフ",
"One repeat, fading.": "一回の繰り返しがフェードします。",
diff --git a/src/crates/concat/locales/ko.json b/src/crates/concat/locales/ko.json
index aef625a0..418f33ed 100644
--- a/src/crates/concat/locales/ko.json
+++ b/src/crates/concat/locales/ko.json
@@ -235,6 +235,7 @@
"Go to end": "끝으로",
"Go to start": "처음으로",
"Green": "녹색",
+ "Generated": "생성됨",
"Green Screen": "그린 스크린",
"Half": "1/2",
"Hall": "홀",
@@ -340,6 +341,7 @@
"Nothing to read yet": "아직 읽을 내용이 없습니다",
"Nothing under the playhead": "재생 헤드 아래에 아무것도 없습니다",
"Nothing was said in that clip": "그 클립에는 말소리가 없습니다",
+ "Nothing read aloud yet. Text to speech puts its voice here.": "아직 읽어 준 것이 없습니다. 텍스트 음성 변환의 음성이 여기에 들어옵니다.",
"Object": "사물",
"off": "끔",
"One repeat, fading.": "한 번 반복하며 사라집니다.",
diff --git a/src/crates/concat/locales/pt-BR.json b/src/crates/concat/locales/pt-BR.json
index cfea8c44..f5d8d432 100644
--- a/src/crates/concat/locales/pt-BR.json
+++ b/src/crates/concat/locales/pt-BR.json
@@ -235,6 +235,7 @@
"Go to end": "Ir para o fim",
"Go to start": "Ir para o início",
"Green": "Verde",
+ "Generated": "Gerado",
"Green Screen": "Tela verde",
"Half": "Metade",
"Hall": "Salão",
@@ -340,6 +341,7 @@
"Nothing to read yet": "Nada para ler ainda",
"Nothing under the playhead": "Nada sob o cursor",
"Nothing was said in that clip": "Nada foi dito nesse clipe",
+ "Nothing read aloud yet. Text to speech puts its voice here.": "Nada foi lido ainda. Texto para fala deixa sua voz aqui.",
"Object": "Objeto",
"off": "desligado",
"One repeat, fading.": "Uma repetição que se esvai.",
diff --git a/src/crates/concat/locales/ru.json b/src/crates/concat/locales/ru.json
index 10ff1ca6..87269464 100644
--- a/src/crates/concat/locales/ru.json
+++ b/src/crates/concat/locales/ru.json
@@ -235,6 +235,7 @@
"Go to end": "В конец",
"Go to start": "В начало",
"Green": "Зелёный",
+ "Generated": "Созданное",
"Green Screen": "Зелёный экран",
"Half": "Половина",
"Hall": "Зал",
@@ -340,6 +341,7 @@
"Nothing to read yet": "Пока нечего читать",
"Nothing under the playhead": "Под курсором ничего нет",
"Nothing was said in that clip": "В этом клипе ничего не сказано",
+ "Nothing read aloud yet. Text to speech puts its voice here.": "Пока ничего не озвучено. Озвучивание текста кладёт свой голос сюда.",
"Object": "Предмет",
"off": "выкл.",
"One repeat, fading.": "Один повтор, затухающий.",
diff --git a/src/crates/concat/locales/tr.json b/src/crates/concat/locales/tr.json
index 28423538..37296d2e 100644
--- a/src/crates/concat/locales/tr.json
+++ b/src/crates/concat/locales/tr.json
@@ -235,6 +235,7 @@
"Go to end": "Sona git",
"Go to start": "Başa git",
"Green": "Yeşil",
+ "Generated": "Oluşturulan",
"Green Screen": "Yeşil perde",
"Half": "Yarım",
"Hall": "Salon",
@@ -340,6 +341,7 @@
"Nothing to read yet": "Henüz okunacak bir şey yok",
"Nothing under the playhead": "Oynatma imlecinin altında bir şey yok",
"Nothing was said in that clip": "O klipte hiçbir şey söylenmedi",
+ "Nothing read aloud yet. Text to speech puts its voice here.": "Henüz hiçbir şey seslendirilmedi. Metinden konuşma sesini buraya bırakır.",
"Object": "Nesne",
"off": "kapalı",
"One repeat, fading.": "Sönen tek bir tekrar.",
diff --git a/src/crates/concat/locales/zh-Hans.json b/src/crates/concat/locales/zh-Hans.json
index 04d917c8..bbafbc7f 100644
--- a/src/crates/concat/locales/zh-Hans.json
+++ b/src/crates/concat/locales/zh-Hans.json
@@ -235,6 +235,7 @@
"Go to end": "跳到结尾",
"Go to start": "跳到开头",
"Green": "绿色",
+ "Generated": "已生成",
"Green Screen": "绿幕",
"Half": "1/2",
"Hall": "大厅",
@@ -340,6 +341,7 @@
"Nothing to read yet": "还没有可朗读的内容",
"Nothing under the playhead": "播放头下方没有内容",
"Nothing was said in that clip": "该片段中没有说话内容",
+ "Nothing read aloud yet. Text to speech puts its voice here.": "尚未朗读任何内容。文字转语音的声音会放在这里。",
"Object": "物体",
"off": "关",
"One repeat, fading.": "一次重复,逐渐消失。",
diff --git a/src/crates/concat/locales/zh-TW.json b/src/crates/concat/locales/zh-TW.json
index f46d362e..96af4b1c 100644
--- a/src/crates/concat/locales/zh-TW.json
+++ b/src/crates/concat/locales/zh-TW.json
@@ -368,6 +368,7 @@
"Generate Captions": "產生字幕",
"Generate captions": "產生字幕",
"Generate speech": "產生語音",
+ "Generated": "已生成",
"Generating voice": "正在產生語音",
"Glitch": "故障",
"Glitch Warp": "故障扭曲",
@@ -559,6 +560,7 @@
"not installed": "未安裝",
"Nothing found to keep here, so the picture shows as shot. Try another subject, or custom removal.": "找不到可保留的主體,因此顯示原始畫面。請嘗試其他主體選項,或使用 [自訂去背]。",
"Nothing of that kind in this project.": "此專案沒有這類內容。",
+ "Nothing read aloud yet. Text to speech puts its voice here.": "尚未朗讀任何內容。文字轉語音的聲音會放在這裡。",
"Nothing starred yet — hover a tile and press its star.": "尚無收藏。請將游標移至卡片上,再按下星號即可收藏。",
"Nothing to caption yet": "尚無可製作字幕的內容",
"Nothing to export": "沒有可匯出的內容",
diff --git a/src/crates/concat/src/panes/media_bin.rs b/src/crates/concat/src/panes/media_bin.rs
index 844747dd..19069762 100644
--- a/src/crates/concat/src/panes/media_bin.rs
+++ b/src/crates/concat/src/panes/media_bin.rs
@@ -16,7 +16,7 @@ use std::path::PathBuf;
use concat_host::media::{self, MediaSummary};
use concat_project::Command;
-use concat_project::model::{self, MediaItem, Project};
+use concat_project::model::{self, MediaItem, MediaOrigin, Project};
use crate::format::wave_path;
use crate::host::{probe_error, spawn};
@@ -211,13 +211,18 @@ impl MediaBin {
project.media_by_id(id)
}
- /// Whether the filter lets an item of this kind through.
- fn shows(filter: MediaFilter, kind: model::MediaKind) -> bool {
+ /// Whether the filter lets an item through. The Media shelves, "All
+ /// media" included, are the imports: a file the editor made is on its
+ /// origin's shelf under Generated and nowhere else, so a read-aloud
+ /// voice is not also the fifth thing under Audio.
+ fn shows(filter: MediaFilter, item: &MediaItem) -> bool {
+ let imported = item.origin.is_none();
match filter {
- MediaFilter::All => true,
- MediaFilter::Video => kind == model::MediaKind::Video,
- MediaFilter::Audio => kind == model::MediaKind::Audio,
- MediaFilter::Images => kind == model::MediaKind::Image,
+ MediaFilter::All => imported,
+ MediaFilter::Video => imported && item.kind == model::MediaKind::Video,
+ MediaFilter::Audio => imported && item.kind == model::MediaKind::Audio,
+ MediaFilter::Images => imported && item.kind == model::MediaKind::Image,
+ MediaFilter::Speech => item.origin == Some(MediaOrigin::Speech),
}
}
@@ -251,7 +256,7 @@ impl MediaBin {
HashSet::new()
};
for item in &project.media {
- if !Self::shows(self.filter, item.kind) {
+ if !Self::shows(self.filter, item) {
continue;
}
let (row, col) = (cell / columns.max(1), cell % columns.max(1));
@@ -268,7 +273,7 @@ impl MediaBin {
let mut visible: Vec<&MediaItem> = project
.media
.iter()
- .filter(|item| Self::shows(self.filter, item.kind))
+ .filter(|item| Self::shows(self.filter, item))
.collect();
match self.sort {
// Added: the import order, as the document keeps it.
@@ -406,6 +411,39 @@ mod tests {
);
}
+ #[test]
+ fn a_generated_voice_is_on_its_own_shelf_and_no_other() {
+ let mut bin = MediaBin::default();
+ let mut project = project();
+ project.media.push(MediaItem {
+ id: "s1".to_owned(),
+ name: "Voice 1.wav".to_owned(),
+ kind: Kind::Audio,
+ origin: Some(MediaOrigin::Speech),
+ ..MediaItem::default()
+ });
+ let names = |bin: &MediaBin| -> Vec<String> {
+ bin.visible(&project)
+ .iter()
+ .map(|item| item.name.clone())
+ .collect()
+ };
+ // Not under All media, not under Audio, though it is audio.
+ bin.filter = MediaFilter::All;
+ assert_eq!(names(&bin).len(), 4, "the imports, and only them");
+ bin.filter = MediaFilter::Audio;
+ assert_eq!(names(&bin), ["alpha.wav"]);
+ bin.filter = MediaFilter::Speech;
+ assert_eq!(names(&bin), ["Voice 1.wav"]);
+ // A marquee on the Speech shelf walks the Speech shelf.
+ let caught = bin.band(&project, 3, (0, 2), (0, 0), false);
+ assert_eq!(caught.len(), 1);
+ assert!(caught.contains("s1"));
+ // Rows are minted for it like any other, so a drag can name it.
+ bin.assign_rows(&project);
+ assert_eq!(bin.by_row(&project, 5).expect("row 5").id, "s1");
+ }
+
#[test]
fn a_click_toggles_only_when_additive() {
let mut bin = MediaBin::default();
diff --git a/src/crates/concat/src/panes/speech.rs b/src/crates/concat/src/panes/speech.rs
index 7e346d20..dd91f2ed 100644
--- a/src/crates/concat/src/panes/speech.rs
+++ b/src/crates/concat/src/panes/speech.rs
@@ -13,7 +13,7 @@ use std::sync::Arc;
use concat_host::media::{self, MediaSummary};
use concat_project::Command;
-use concat_project::model::ClipKind;
+use concat_project::model::{ClipKind, MediaOrigin};
use concat_speech::tts::{CHATTERBOX_CLONE, Family, Reference, VoiceInfo, family_of, is_clone};
use slint::SharedString;
@@ -137,9 +137,11 @@ impl SpeechPane {
self.running = false;
match *result {
Ok(summary) => {
- let created = studio.apply(Command::AddMedia {
- item: summary.to_new_media(),
- });
+ // Marked as read aloud: the bin shelves it under
+ // Generated › Speech, apart from the imports.
+ let mut item = summary.to_new_media();
+ item.origin = Some(MediaOrigin::Speech);
+ let created = studio.apply(Command::AddMedia { item });
let media_id = created.or_else(|| {
studio
.project()
diff --git a/src/crates/concat/src/studio.rs b/src/crates/concat/src/studio.rs
index 08230730..6f4e4478 100644
--- a/src/crates/concat/src/studio.rs
+++ b/src/crates/concat/src/studio.rs
@@ -6403,25 +6403,24 @@ impl Studio {
);
// The bin.
+ // The Media shelves count the imports; what the editor made is
+ // counted on its own shelf under Generated. See MediaBin::shows.
let items = &self.project().media;
sync(&models.media, self.media.rows(self));
- editor.set_media_count_all(items.len() as i32);
- editor.set_media_count_video(
+ let imported = |kind: Option<model::MediaKind>| {
items
.iter()
- .filter(|item| item.kind == model::MediaKind::Video)
- .count() as i32,
- );
- editor.set_media_count_audio(
- items
- .iter()
- .filter(|item| item.kind == model::MediaKind::Audio)
- .count() as i32,
- );
- editor.set_media_count_images(
+ .filter(|item| item.origin.is_none() && kind.is_none_or(|kind| item.kind == kind))
+ .count() as i32
+ };
+ editor.set_media_count_all(imported(None));
+ editor.set_media_count_video(imported(Some(model::MediaKind::Video)));
+ editor.set_media_count_audio(imported(Some(model::MediaKind::Audio)));
+ editor.set_media_count_images(imported(Some(model::MediaKind::Image)));
+ editor.set_media_count_speech(
items
.iter()
- .filter(|item| item.kind == model::MediaKind::Image)
+ .filter(|item| item.origin == Some(model::MediaOrigin::Speech))
.count() as i32,
);
editor.set_media_selected_count(self.media.selected.len() as i32);
diff --git a/src/crates/concat/ui/editor.slint b/src/crates/concat/ui/editor.slint
index 92b4108c..5d1a4db0 100644
--- a/src/crates/concat/ui/editor.slint
+++ b/src/crates/concat/ui/editor.slint
@@ -70,6 +70,7 @@ export global Editor {
in-out property <int> media-count-video;
in-out property <int> media-count-audio;
in-out property <int> media-count-images;
+ in-out property <int> media-count-speech;
in-out property <int> media-selected-count;
in-out property <bool> importing;
diff --git a/src/crates/concat/ui/workspace/media-pane.slint b/src/crates/concat/ui/workspace/media-pane.slint
index ed62bd99..13d56c24 100644
--- a/src/crates/concat/ui/workspace/media-pane.slint
+++ b/src/crates/concat/ui/workspace/media-pane.slint
@@ -20,9 +20,10 @@ import { Wave } from "../primitives/wave.slint";
export enum MediaKind { video, audio, image }
/// What the sidebar has narrowed the bin to. `all` is not "no filter": it is
-/// the row that is highlighted when every kind is on, which is why it is a
-/// value here and not the absence of one.
-export enum MediaFilter { all, video, audio, images }
+/// the row that is highlighted when every import is on, which is why it is a
+/// value here and not the absence of one. `speech` is the Generated group's
+/// shelf: what the editor read aloud, which the Media rows leave out.
+export enum MediaFilter { all, video, audio, images, speech }
/// One look a title can be given, as the Text page's Presets shelf shows
/// it: enough of the style to draw the name in it. Published by Rust from
@@ -1356,6 +1357,8 @@ export component MediaPane inherits Rectangle {
in property <int> count-video;
in property <int> count-audio;
in property <int> count-images;
+ /// What the editor made, by origin; not in `count-all`.
+ in property <int> count-speech;
/// how many of the whole bin are selected, not how many of `items` are
in property <int> selected-count;
in property <bool> importing;
@@ -1422,6 +1425,7 @@ export component MediaPane inherits Rectangle {
property <int> tab;
property <bool> categories-open: true;
+ property <bool> generated-open: true;
property <bool> text-open: true;
/// 0 = Titles, 1 = Presets
property <int> text-category;
@@ -1805,6 +1809,25 @@ export component MediaPane inherits Rectangle {
current: root.filter == MediaFilter.images;
clicked => { root.narrow(MediaFilter.images); }
}
+
+ // What the editor made rather than the user
+ // imported, one shelf per origin. A voice read
+ // aloud is here and not under Audio, so the sound
+ // the user brought stays tellable from the sound
+ // the app wrote.
+ CategoryHeading {
+ height: 26px;
+ title: I18n.t("Generated");
+ open: root.generated-open;
+ toggled => { root.generated-open = !root.generated-open; }
+ }
+ if root.generated-open: CategoryRow {
+ height: 24px;
+ label: I18n.t("Speech");
+ count: root.count-speech;
+ current: root.filter == MediaFilter.speech;
+ clicked => { root.narrow(MediaFilter.speech); }
+ }
}
if root.tab == 1: VerticalLayout {
@@ -2026,7 +2049,9 @@ export component MediaPane inherits Rectangle {
if root.tab == 0 && root.items.length == 0: Empty {
glyph: Glyph.import;
- text: root.count-all > 0
+ text: root.filter == MediaFilter.speech
+ ? I18n.t("Nothing read aloud yet. Text to speech puts its voice here.")
+ : root.count-all > 0
? I18n.t("Nothing of that kind in this project.")
: "Drop video and audio files anywhere in the window to import them.";
}
diff --git a/src/crates/concat/ui/workspace/seat.slint b/src/crates/concat/ui/workspace/seat.slint
index a6413e0b..33ca262e 100644
--- a/src/crates/concat/ui/workspace/seat.slint
+++ b/src/crates/concat/ui/workspace/seat.slint
@@ -97,6 +97,7 @@ export component Seat inherits Rectangle {
count-video: Editor.media-count-video;
count-audio: Editor.media-count-audio;
count-images: Editor.media-count-images;
+ count-speech: Editor.media-count-speech;
selected-count: Editor.media-selected-count;
importing: Editor.importing;