Update app.py

Update app.py

Update app.py

Update UI to modern Dark Mode and add Speed/Split params

Switch to YourTTS (VITS) for speed

Downgrade transformers to 4.38.2 and add get_logits_warper monkeypatch

Enhance torch.isin monkeypatch for robust int handling

Monkeypatch torch.isin to fix transformers compatibility

Add root redirect to UI

Downgrade transformers to 4.42.0 for compatibility

Fix dependencies: Downgrade transformers to 4.46.0 and monkeypatch isin bug

Fix model download: Use huggingface_hub instead of Coqui gateway

Upgrade transformers to fix streaming bug

Fix TTS model attribute access for streaming

Fix torch.load monkeypatch order

Add UI and extended API parameters

Docs: Update API guide with PyTorch 2.6+ compatibility notes and streaming examples

Fix: Monkeypatch torch.load to allow legacy Coqui checkpoints on PyTorch 2.6+

Fix: Pin torch to 2.4.0 to prevent weights_only load error

Fix: Upgrade transformers to 4.46.0 to resolve torch.isin compatibility issue

Docs: Update for Hugging Face Spaces deployment

Docs: Add streaming code example

Debug: Add logging for streaming chunks and update docs

Fix: Remove enable_text_splitting and debug empty generator

Debug: Yield error message in stream

Fix audio streaming: enable text splitting and chunk size

Fix: Use tts.synthesizer.tts_model for streaming

Fix: Remove unsupported parameters from inference_stream

Add debugging for streaming errors

Add /debug_tts endpoint

Add Voice Management Endpoints (Upload/Delete/Rename)

Add technical explanation of streaming architecture

Add /stream endpoint and generation parameters

Update docs with full voice list

Add /voices endpoint and update docs

Add new voices (MP3) and update gitattributes

Fix: Downgrade Torch to 2.4.0 to avoid PyTorch 2.6 weights_only=True breaking change

Fix: Pin transformers version to 4.41.2 to resolve BeamSearchScorer import error

Update speakers: Add real voice samples (Jiayan)

Debug: Add /status endpoint and capture load error

Update API: Add speakers directory, LFS tracking, and /speakers endpoint

Deploy XTTS-v2 Multilingual API (Fixed Metadata)

initial commit