wangyuandHuang, Zeyu
6f2ad7b403
Signed-off-by: wangyu <410167048@qq.com> Signed-off-by: Huang, Zeyu <11222265+fhfuih@users.noreply.github.com> Signed-off-by: wangyu <53896905+yenuo26@users.noreply.github.com> Co-authored-by: Huang, Zeyu <11222265+fhfuih@users.noreply.github.com>
20 lines
594 B
YAML
20 lines
594 B
YAML
# OmniVoice TTS deploy: diffusion engine single stage (dit).
|
|
# Text → iterative unmasking → codebook tokens → DAC decode → 24kHz audio.
|
|
# (Migrated from model_executor/stage_configs/omnivoice.yaml for deploy/ layout.)
|
|
|
|
stage_args:
|
|
- stage_id: 0
|
|
stage_type: diffusion
|
|
is_comprehension: true
|
|
runtime:
|
|
devices: 0
|
|
engine_args:
|
|
model_stage: dit
|
|
model_class_name: "OmniVoicePipeline"
|
|
enforce_eager: true
|
|
trust_remote_code: true
|
|
distributed_executor_backend: "mp"
|
|
dtype: "float32"
|
|
final_output: true
|
|
final_output_type: audio
|