Pascal 7f532f1ed0 cli, pipeline: decode --ref-audio once at the CLI, pipeline takes a mono buffer
Clone mode used to read the same WAV twice (speaker encoder then codec
encoder), which failed silently on Windows where pipeline_codec_encode
returned empty on the second pass. PipelineTTSSynthesizeParams now takes
ref_audio_24k + ref_n_samples instead of a path. The CLI calls
audio_read_mono once, holds the buffer in a unique_ptr until synthesis
returns, and feeds the same pointer to both encoders.
2026-05-11 18:14:59 +02:00
2026-05-10 20:56:28 +02:00
2026-05-11 16:27:12 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
2026-05-10 15:57:15 +02:00
S
Description
qwentts.cpp fork: tts-server Qwen3-TTS con containerizzazione Vulkan
MIT
2.9 MiB
Languages
C++ 81.6%
Python 13.1%
C 1.6%
Shell 1.4%
CMake 1.2%
Other 1.1%