Skip to content

The open-source AI voice studio. Clone, dictate, create.

Regenerate as new version, UI polish, and bugfixes

Jamie Pine00c5b75ffb7c
diff --git a/app/src/components/AudioPlayer/AudioPlayer.tsx b/app/src/components/AudioPlayer/AudioPlayer.tsx
index 44ebb266..75e1b4e5 100644
--- a/app/src/components/AudioPlayer/AudioPlayer.tsx
+++ b/app/src/components/AudioPlayer/AudioPlayer.tsx
@@ -157,8 +157,21 @@ export function AudioPlayer() {
const wavesurfer = wavesurferRef.current;
if (!wavesurfer) return;
- // Update store when time changes
+ // Update store when time changes, stop if past duration
wavesurfer.on('timeupdate', (time) => {
+ const dur = usePlayerStore.getState().duration;
+ if (dur > 0 && time >= dur) {
+ setCurrentTime(dur);
+ const loop = usePlayerStore.getState().isLooping;
+ if (loop) {
+ wavesurfer.seekTo(0);
+ wavesurfer.play();
+ } else {
+ wavesurfer.pause();
+ setIsPlaying(false);
+ }
+ return;
+ }
setCurrentTime(time);
});
diff --git a/app/src/components/History/HistoryTable.tsx b/app/src/components/History/HistoryTable.tsx
index 72a1fc90..7900cd44 100644
--- a/app/src/components/History/HistoryTable.tsx
+++ b/app/src/components/History/HistoryTable.tsx
@@ -224,6 +224,20 @@ export function HistoryTable() {
}
};
+ const handleRegenerate = async (generationId: string) => {
+ try {
+ await apiClient.regenerateGeneration(generationId);
+ addPendingGeneration(generationId);
+ queryClient.invalidateQueries({ queryKey: ['history'] });
+ } catch (error) {
+ toast({
+ title: 'Regenerate failed',
+ description: error instanceof Error ? error.message : 'Could not regenerate',
+ variant: 'destructive',
+ });
+ }
+ };
+
const handleApplyEffects = (generationId: string) => {
setEffectsTargetId(generationId);
setEffectsChain([]);
@@ -457,7 +471,7 @@ export function HistoryTable() {
>
<RotateCcw className="h-4 w-4" />
</Button>
- ) : isPlayable ? (
+ ) : (
<>
<DropdownMenu>
<DropdownMenuTrigger asChild>
@@ -466,6 +480,7 @@ export function HistoryTable() {
size="icon"
className="h-8 w-8"
aria-label="Actions"
+ disabled={isGenerating}
>
<MoreHorizontal className="h-4 w-4" />
</Button>
@@ -495,6 +510,10 @@ export function HistoryTable() {
<Wand2 className="mr-2 h-4 w-4" />
Apply Effects
</DropdownMenuItem>
+ <DropdownMenuItem onClick={() => handleRegenerate(gen.id)}>
+ <RotateCcw className="mr-2 h-4 w-4" />
+ Regenerate
+ </DropdownMenuItem>
<DropdownMenuItem
onClick={() => handleDeleteClick(gen.id, gen.profile_name)}
disabled={deleteGeneration.isPending}
@@ -519,7 +538,7 @@ export function HistoryTable() {
</Button>
)}
</>
- ) : null}
+ )}
</div>
</div>
@@ -533,13 +552,13 @@ export function HistoryTable() {
transition={{ duration: 0.2, ease: 'easeOut' }}
className="overflow-hidden"
>
- <div className="border-t border-border/50 px-3 pb-2 pt-2">
+ <div className="border-t border-border/50">
<div className="divide-y divide-border/40">
{gen.versions.map((v) => (
<button
key={v.id}
type="button"
- className="flex items-center gap-2 w-full h-9 px-2 text-left hover:bg-muted/50 transition-colors"
+ className="flex items-center gap-2 w-full h-9 px-3 text-left hover:bg-muted/50 transition-colors"
onClick={() => {
handlePlayVersion(gen.id, v.id, gen.text, gen.profile_id);
if (!v.is_default) {
@@ -550,8 +569,8 @@ export function HistoryTable() {
<Play className="h-3 w-3 shrink-0 text-muted-foreground" />
<span className="truncate text-xs font-medium">{v.label}</span>
{v.effects_chain && v.effects_chain.length > 0 && (
- <span className="text-[10px] text-muted-foreground">
- {v.effects_chain.length} fx
+ <span className="text-[10px] text-muted-foreground truncate">
+ {v.effects_chain.map((e) => e.type).join(' → ')}
</span>
)}
<span className="flex-1" />
diff --git a/app/src/lib/api/client.ts b/app/src/lib/api/client.ts
index fd3e1b50..1310c383 100644
--- a/app/src/lib/api/client.ts
+++ b/app/src/lib/api/client.ts
@@ -212,6 +212,12 @@ class ApiClient {
});
}
+ async regenerateGeneration(generationId: string): Promise<GenerationResponse> {
+ return this.request<GenerationResponse>(`/generate/${generationId}/regenerate`, {
+ method: 'POST',
+ });
+ }
+
// History
async listHistory(query?: HistoryQuery): Promise<HistoryListResponse> {
const params = new URLSearchParams();
diff --git a/backend/export_import.py b/backend/export_import.py
index 6d705aa4..7af1f9b0 100644
--- a/backend/export_import.py
+++ b/backend/export_import.py
@@ -13,7 +13,7 @@ from typing import Optional
from sqlalchemy.orm import Session
from .models import VoiceProfileResponse
-from .database import VoiceProfile as DBVoiceProfile, ProfileSample as DBProfileSample, Generation as DBGeneration
+from .database import VoiceProfile as DBVoiceProfile, ProfileSample as DBProfileSample, Generation as DBGeneration, GenerationVersion as DBGenerationVersion
from .profiles import create_profile, add_profile_sample
from .models import VoiceProfileCreate
from . import config
@@ -269,16 +269,33 @@ def export_generation_to_zip(generation_id: str, db: Session) -> bytes:
if not profile:
raise ValueError(f"Profile {generation.profile_id} not found")
- # Get audio file
- audio_path = Path(generation.audio_path)
- if not audio_path.exists():
- raise ValueError(f"Audio file not found: {audio_path}")
-
+ # Get all versions for this generation
+ versions = (
+ db.query(DBGenerationVersion)
+ .filter_by(generation_id=generation_id)
+ .order_by(DBGenerationVersion.created_at)
+ .all()
+ )
+
# Create ZIP in memory
zip_buffer = io.BytesIO()
with zipfile.ZipFile(zip_buffer, 'w', zipfile.ZIP_DEFLATED) as zip_file:
- # Create manifest.json
+ # Build version manifest entries
+ version_entries = []
+ for v in versions:
+ v_path = Path(v.audio_path)
+ effects_chain = None
+ if v.effects_chain:
+ effects_chain = json.loads(v.effects_chain)
+ version_entries.append({
+ "id": v.id,
+ "label": v.label,
+ "is_default": v.is_default,
+ "effects_chain": effects_chain,
+ "filename": v_path.name,
+ })
+
manifest = {
"version": "1.0",
"generation": {
@@ -295,13 +312,22 @@ def export_generation_to_zip(generation_id: str, db: Session) -> bytes:
"name": profile.name,
"description": profile.description,
"language": profile.language,
- }
+ },
+ "versions": version_entries,
}
zip_file.writestr("manifest.json", json.dumps(manifest, indent=2))
- # Add audio file
- filename = audio_path.name
- zip_file.write(audio_path, f"audio/{filename}")
+ # Add all version audio files
+ for v in versions:
+ v_path = Path(v.audio_path)
+ if v_path.exists():
+ zip_file.write(v_path, f"audio/{v_path.name}")
+
+ # Fallback: if no versions exist, include the generation's main audio
+ if not versions:
+ audio_path = Path(generation.audio_path)
+ if audio_path.exists():
+ zip_file.write(audio_path, f"audio/{audio_path.name}")
zip_buffer.seek(0)
return zip_buffer.read()
diff --git a/backend/main.py b/backend/main.py
index e269b20d..31ca5ad9 100644
--- a/backend/main.py
+++ b/backend/main.py
@@ -828,7 +828,7 @@ async def generate_speech(
# Create clean version entry
versions_mod.create_version(
generation_id=generation_id,
- label="clean",
+ label="original",
audio_path=str(clean_audio_path),
db=bg_db,
effects_chain=None,
@@ -849,7 +849,7 @@ async def generate_speech(
final_audio_path = str(processed_path)
versions_mod.create_version(
generation_id=generation_id,
- label="processed",
+ label="version-2",
audio_path=str(processed_path),
db=bg_db,
effects_chain=effects_chain_config,
@@ -978,6 +978,118 @@ async def retry_generation(generation_id: str, db: Session = Depends(get_db)):
return models.GenerationResponse.model_validate(gen)
+ "/generate/{generation_id}/regenerate",
+ response_model=models.GenerationResponse,
+)
+async def regenerate_generation(generation_id: str, db: Session = Depends(get_db)):
+ """Re-run TTS with the same parameters and save the result as a new version."""
+ gen = db.query(DBGeneration).filter_by(id=generation_id).first()
+ if not gen:
+ raise HTTPException(status_code=404, detail="Generation not found")
+ if (gen.status or "completed") != "completed":
+ raise HTTPException(status_code=400, detail="Generation must be completed to regenerate")
+
+ from .backends import get_tts_backend_for_engine
+ from . import versions as versions_mod
+
+ regen_engine = gen.engine or "qwen"
+ regen_model_size = gen.model_size or "1.7B"
+ tts_model = get_tts_backend_for_engine(regen_engine)
+
+ # Set to generating so the UI shows the loader and SSE picks it up
+ gen.status = "generating"
+ gen.error = None
+ db.commit()
+ db.refresh(gen)
+
+ task_manager = get_task_manager()
+ task_manager.start_generation(
+ task_id=generation_id,
+ profile_id=gen.profile_id,
+ text=gen.text,
+ )
+
+ version_id = str(uuid.uuid4())
+
+ async def _run_regenerate():
+ bg_db = next(get_db())
+ try:
+ if regen_engine == "qwen":
+ await tts_model.load_model_async(regen_model_size)
+ else:
+ await tts_model.load_model()
+
+ voice_prompt = await profiles.create_voice_prompt_for_profile(
+ gen.profile_id,
+ bg_db,
+ use_cache=True,
+ engine=regen_engine,
+ )
+
+ from .utils.chunked_tts import generate_chunked
+
+ trim_fn = None
+ if regen_engine in ("chatterbox", "chatterbox_turbo"):
+ from .utils.audio import trim_tts_output
+ trim_fn = trim_tts_output
+
+ audio, sample_rate = await generate_chunked(
+ tts_model,
+ gen.text,
+ voice_prompt,
+ language=gen.language,
+ seed=None, # New seed for variation
+ instruct=gen.instruct,
+ trim_fn=trim_fn,
+ )
+
+ from .utils.audio import normalize_audio, save_audio
+ audio = normalize_audio(audio)
+
+ duration = len(audio) / sample_rate
+ audio_path = config.get_generations_dir() / f"{generation_id}_{version_id[:8]}.wav"
+
+ save_audio(audio, str(audio_path), sample_rate)
+
+ # Count existing versions to auto-label
+ existing = versions_mod.list_versions(generation_id, bg_db)
+ label = f"take-{len(existing) + 1}"
+
+ versions_mod.create_version(
+ generation_id=generation_id,
+ label=label,
+ audio_path=str(audio_path),
+ db=bg_db,
+ effects_chain=None,
+ is_default=True,
+ )
+
+ await history.update_generation_status(
+ generation_id=generation_id,
+ status="completed",
+ db=bg_db,
+ audio_path=str(audio_path),
+ duration=duration,
+ )
+ except Exception as e:
+ import traceback
+ traceback.print_exc()
+ await history.update_generation_status(
+ generation_id=generation_id,
+ status="failed",
+ db=bg_db,
+ error=str(e),
+ )
+ finally:
+ task_manager.complete_generation(generation_id)
+ bg_db.close()
+
+ _enqueue_generation(_run_regenerate())
+
+ return models.GenerationResponse.model_validate(gen)
+
+
@app.get("/generate/{generation_id}/status")
async def get_generation_status(generation_id: str, db: Session = Depends(get_db)):
"""SSE endpoint that streams generation status updates.
@@ -1599,7 +1711,7 @@ async def preview_effects(
# Write to in-memory buffer
import soundfile as sf
buf = io.BytesIO()
- await asyncio.to_thread(sf.write, buf, processed, sample_rate, "WAV")
+ await asyncio.to_thread(lambda: sf.write(buf, processed, sample_rate, format="WAV"))
buf.seek(0)
return StreamingResponse(