0.4.2

GargantuaX/gemini-watermark-remover0.4.2May 19, 2025by edwko

AI Summary

Enhanced OuteTTS with fade-in/fade-out audio decoding, batched decoding interfaces for high-throughput processing, and support for the new OuteTTS 1.0 0.6B model.

Key Highlights

Fade-in/fade-out audio decoding to eliminate clipping artifacts
Batched decoding interfaces: EXL2 Async, VLLM, and llama.cpp Async
Single-stream decoding endpoint for llama.cpp server
OuteTTS 1.0 0.6B model support
New batched interface configuration parameters

New Features

Audio fade effects
Batched inference support
Async processing backends
New model compatibility
Enhanced pre-prompt normalization

Full Release Notes

# OuteTTS v0.4.2

* **Fade-in / Fade-out Audio Decoding**
  Introduced quick fade-in and fade-out on decoded audio chunks to eliminate clipping artifacts at segment boundaries.

* **Batched Decoding Interfaces**
  Added support for high-throughput, batched inference via three new backends:

  * **EXL2 Async**: Asynchronous batch processing using the EXL2.
  * **VLLM**: Asynchronous batch decoding with VLLM. (Experiment support)
  * **llama.cpp Async Server Endpoint**: Connects to a continuously-batched llama.cpp server for async inference.

* **Single-Stream Decoding**

  * **llama.cpp Server Endpoint**: single-stream decode endpoint for llama.cpp server.

* **OuteTTS 1.0 0.6B Model Support**
  Compatibility with the new [OuteTTS-1.0-0.6B](https://huggingface.co/OuteAI/OuteTTS-1.0-0.6B), including config defaults.

* **Batched Interface Parameters**
  New configuration options to control batched interface.

* Enhanced pre-prompt normalization pipeline.

* Documentation Updates, expanded the batched interface usage.