mARTin-B78/tts-dgx-spark-faster-qwen3-tts
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
42 Commits 1 Branch 0 Tags 323 KiB
9836261d28
Commit Graph

4 Commits

Author SHA1 Message Date
mARTin-B78
3c389b0292 v6.7: Native speed control and zip word-level timestamps 2026-06-26 11:11:13 +02:00
mARTin-B78
79d8d2cd96 fix(customvoice): override max_seq_len to 1024 to prevent CUDA OOM 2026-06-22 11:17:39 +02:00
mARTin-B78
d04fa88853 refactor: load models asynchronously and add CUDA warmup for VoiceDesign and CustomVoice servers 2026-05-26 23:07:51 +02:00
mARTin-B78
8363dc7eb1 Add CustomVoice server, benchmark tool, and VoiceDesign API improvements 
- Add run_customvoice_server.py: OpenAI-compatible TTS server for Qwen3-TTS
  CustomVoice models with speaker-based voice selection and per-request
  language/instruct/max_new_tokens overrides
- Add customvoice_voices.json: voice config for 9 built-in speakers across
  English, Chinese, Japanese, and Korean
- Add benchmark_api.py: API benchmarking tool reporting TTFA, total time,
  RTF, and speed multiplier across short/medium/long sentences
- Refactor run_voicedesign_server.py: extract _request_generation_params()
  helper, add per-request language/instruct/max_new_tokens override support,
  and wire DEFAULT_MAX_NEW_TOKENS to --max-seq-len arg
- Add faster-qwen3-tts.code-workspace for VS Code project config

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-05-26 12:19:33 +02:00