Add optional 44.1 kHz X-VC enhancement

This commit is contained in:
Mikei386 committed 2026-09-09 17:21:20 +02:00
1 parent b4f4bf37fd
commit 51ed201f6c
9 files changed
+128 -14

No files matched your search

+7 -1
View File
@@ -9,7 +9,7 @@ ZeroGPU.
- Private URL: `http://192.168.1.212:8009`
- Source clip: speech content and timing to preserve
- Reference clip: target speaker identity
- Output: 16 kHz PCM WAV
- Output: native 16 kHz PCM WAV plus optional Resemble-Enhance restoration at 44.1 kHz
- GPU: RTX 5080 only
- Persistent cache: `/data/voice/xvc/huggingface`
- Code and model license: MIT
@@ -22,3 +22,9 @@ Technical acceptance on 9 September 2026 used the repository's source and
target examples: 5.20 seconds were converted in 1.07 seconds (RTF 0.21). The
result was valid mono PCM WAV at 16 kHz, and the loaded process occupied about
2.9 GiB on the RTX 5080. German listening quality remains open.
The optional high-quality path uses `resemble-enhance` 0.0.1 with model
revision `4e3510ce4a8391159f665903544c5150bee7b2cb`. It does not change X-VC's
native 16-kHz architecture. Instead, it reconstructs missing speech bandwidth
after conversion and writes a second 44.1-kHz WAV. The UI always retains the
native output for an honest A/B comparison.