[Bug]Higgs Audio SRT processing failed: 'CharacterParser' object has no attribute 'get_language_display_name'

[WLimin]

Describe the bug
I'm using the example workflow 'Unified 📺 TTS SRT.json' with the latest git code, and run Higgs Audio 2 Engine with CPU.
A error message is reported:
`raw

2026-04-12T01:35:03.303126220Z 🎤 Generating Higgs Audio for 1 chunk(s)
2026-04-12T01:35:03.303130373Z Processing chunk 1/1: 43 chars / ~10 tokens
2026-04-12T01:35:03.303131663Z Using reference audio for voice cloning
2026-04-12T01:35:03.307287102Z 🗣️ Generating audio...
2026-04-12T01:35:07.929994447Z 🧹 Cleared DynamicCache state for fresh generation
2026-04-12T01:35:07.930025010Z 🧹 Cleared DynamicCache state for fresh generation
2026-04-12T01:35:07.930026683Z 🧹 Cleared DynamicCache state for fresh generation
🗣️ Generating: 2%|██████▋ 2026-04-12T01:36:16.542656932Z | 45/20🗣️ Generating: 4%|████████████▉ 2026-04-12T01:36:16.542656932Z | 87/20❌ Higgs Audio SRT processing failed: 'CharacterParser' object has no attribute 'get_language_display_name'
2026-04-12T01:36:55.555612617Z Traceback (most recent call last):
2026-04-12T01:36:55.555613989Z File "/data/config/comfy/custom_nodes/tts_audio_suite/nodes/higgs_audio/higgs_audio_srt_processor.py", line 308, in generate_srt_speech
2026-04-12T01:36:55.555614905Z result = pause_processor.generate_audio_with_pauses(
2026-04-12T01:36:55.555615467Z ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2026-04-12T01:36:55.555616067Z File "/data/config/comfy/custom_nodes/tts_audio_suite/utils/text/pause_processor.py", line 126, in generate_audio_with_pauses
2026-04-12T01:36:55.555616671Z audio = tts_generate_func(content, **generation_kwargs)
2026-04-12T01:36:55.555617261Z ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2026-04-12T01:36:55.555617830Z File "/data/config/comfy/custom_nodes/tts_audio_suite/nodes/higgs_audio/higgs_audio_srt_processor.py", line 235, in srt_tts_generate_func
2026-04-12T01:36:55.555618539Z display_name = character_parser.get_language_display_name(language)
2026-04-12T01:36:55.555619218Z ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2026-04-12T01:36:55.555619816Z AttributeError: 'CharacterParser' object has no attribute 'get_language_display_name'
`

I use the latest git repository code. So, I greped the code:
`raw

/comfy/custom_nodes/tts_audio_suite/utils/text$ grep -Ern 'split_by_character_with_language'
character_parser/base_parser.py:277: def split_by_character_with_language(self, text: str) -> List[Tuple[str, str, str]]:
character_parser/base_parser.py:291: def split_by_character_with_language_and_explicit_flag(self, text: str) -> List[Tuple[str, str, str, bool]]:
character_parser/base_parser.py:320: split_by_character_with_language() and will never see emotion data.
character_parser_backup.py:740: def split_by_character_with_language(self, text: str) -> List[Tuple[str, str, str]]:
character_parser_backup.py:754: def split_by_character_with_language_and_explicit_flag(self, text: str) -> List[Tuple[str, str, str, bool]]:
/comfy/custom_nodes/tts_audio_suite/utils/text$ grep -Ern 'get_language_display_name'
character_parser/validation.py:182: display_name = language_resolver.get_language_display_name(canonical_lang)
character_parser_backup.py:667: def get_language_display_name(self, language_code: str) -> str:
`

I think that is a temporary issues, so I modified the file "/comfy/custom_nodes/tts_audio_suite/nodes/higgs_audio/higgs_audio_srt_processor.py", line 235, from
`python

                            if explicit_language and language:
                                display_name = character_parser.get_language_display_name(language)
                                higgs_text = f"[{display_name}] {segment_text}"

topython

                            if explicit_language and language:
                                from utils.text.character_parser_backup import character_parser as character_parser_backup
                                display_name = character_parser_backup.get_language_display_name(language)
                                higgs_text = f"[{display_name}] {segment_text}"

`
I think it works fine.

[diodiogod]

Tentative fix pushed in v4.25.12 on main.

What changed:

• Restored the missing language display helper used by Higgs Audio SRT character/language tag handling
• Added a compatibility shim for the newer upstream Higgs Hugging Face config layout
• Added clearer Higgs config detection logs to help distinguish legacy vs newer upstream schema

Please update to the latest main and test again.

For this issue specifically, the expected result is that Higgs Audio SRT no longer fails with:
'CharacterParser' object has no attribute 'get_language_display_name'

If it still fails, please paste the new full log.