Skip to content

higgs tts input异常 #140

Description

@paleyoupa

python examples/higgs_audio_v3_quickstart.py
--checkpoint /workspace/models/higgs-audio-v3-tts-4b
--text "Four score."
--out test_four_score.wav
Could not determine semantic model type from config architecture. Defaulting to HubertConfig.
[FP8 W8A8] 'Four score.'
-> test_four_score.wav (40.68s audio, 17.80s wall incl 1st-call setup)
done.
用"Four score."生成的wav文件也是杂音,40.68s应该是超时了

python examples/higgs_audio_v3_quickstart.py
--checkpoint /workspace/models/higgs-audio-v3-tts-4b
--text "Four years ago."
--out test_four_years.wav
Could not determine semantic model type from config architecture. Defaulting to HubertConfig.
[FP8 W8A8] 'Four years ago.'
-> test_four_years.wav (1.32s audio, 7.28s wall incl 1st-call setup)
done.
用其他生成没问题

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

No type

Fields

No fields configured for issues without a type.

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions