Can't import transformer packages

[zhongli1990]

I am running it on AWS Ubuntu T4, but the docker compose failed with below error message?

realtime-voice-chat-app | INFO: Waiting for application startup.
realtime-voice-chat-app | 24:57.16 server INFO 🖥️▶️ Server starting up
realtime-voice-chat-app | 24:57.16 root INFO Local XTTS Model: "Lasinya" specified
realtime-voice-chat-app | 24:57.16 root INFO Local models path: "./models/Lasinya"
realtime-voice-chat-app | 24:57.16 root INFO config.json exists in ./models/Lasinya/config.json (no download).
realtime-voice-chat-app | 24:57.16 root INFO model.pth exists in ./models/Lasinya/model.pth (no download).
realtime-voice-chat-app | 24:57.16 root INFO vocab.json exists in ./models/Lasinya/vocab.json (no download).
realtime-voice-chat-app | 24:57.16 root INFO speakers_xtts.pth exists in ./models/Lasinya/speakers_xtts.pth (no download).
realtime-voice-chat-app | Start method has already been set. Details: context has already been set
realtime-voice-chat-app | ERROR: Traceback (most recent call last):
realtime-voice-chat-app | File "/usr/local/lib/python3.10/dist-packages/starlette/routing.py", line 694, in lifespan
realtime-voice-chat-app | async with self.lifespan_context(app) as maybe_state:
realtime-voice-chat-app | File "/usr/lib/python3.10/contextlib.py", line 199, in aenter
realtime-voice-chat-app | return await anext(self.gen)
realtime-voice-chat-app | File "/app/code/server.py", line 119, in lifespan
realtime-voice-chat-app | app.state.SpeechPipelineManager = SpeechPipelineManager(
realtime-voice-chat-app | File "/app/code/speech_pipeline_manager.py", line 155, in init
realtime-voice-chat-app | self.audio = AudioProcessor(
realtime-voice-chat-app | File "/app/code/audio_module.py", line 107, in init
realtime-voice-chat-app | self.engine = CoquiEngine(
realtime-voice-chat-app | File "/usr/local/lib/python3.10/dist-packages/RealtimeTTS/engines/base_engine.py", line 36, in call
realtime-voice-chat-app | instance.post_init()
realtime-voice-chat-app | File "/usr/local/lib/python3.10/dist-packages/RealtimeTTS/engines/coqui_engine.py", line 215, in post_init
realtime-voice-chat-app | self.create_worker_process()
realtime-voice-chat-app | File "/usr/local/lib/python3.10/dist-packages/RealtimeTTS/engines/coqui_engine.py", line 243, in create_worker_process
realtime-voice-chat-app | self.retrieve_coqui_voices()
realtime-voice-chat-app | File "/usr/local/lib/python3.10/dist-packages/RealtimeTTS/engines/coqui_engine.py", line 1146, in retrieve_coqui_voices
realtime-voice-chat-app | from TTS.tts.layers.xtts.xtts_manager import SpeakerManager
realtime-voice-chat-app | File "/usr/local/lib/python3.10/dist-packages/TTS/init.py", line 25, in
realtime-voice-chat-app | from TTS.tts.configs.xtts_config import XttsConfig
realtime-voice-chat-app | File "/usr/local/lib/python3.10/dist-packages/TTS/tts/configs/xtts_config.py", line 4, in
realtime-voice-chat-app | from TTS.tts.models.xtts import XttsArgs, XttsAudioConfig
realtime-voice-chat-app | File "/usr/local/lib/python3.10/dist-packages/TTS/tts/models/xtts.py", line 13, in
realtime-voice-chat-app | from TTS.tts.layers.xtts.gpt import GPT
realtime-voice-chat-app | File "/usr/local/lib/python3.10/dist-packages/TTS/tts/layers/xtts/gpt.py", line 16, in
realtime-voice-chat-app | from TTS.tts.layers.xtts.gpt_inference import GPT2InferenceModel
realtime-voice-chat-app | File "/usr/local/lib/python3.10/dist-packages/TTS/tts/layers/xtts/gpt_inference.py", line 6, in
realtime-voice-chat-app | from TTS.tts.layers.xtts.stream_generator import StreamGenerationConfig
realtime-voice-chat-app | File "/usr/local/lib/python3.10/dist-packages/TTS/tts/layers/xtts/stream_generator.py", line 13, in
realtime-voice-chat-app | from transformers import (
realtime-voice-chat-app | ImportError: cannot import name 'BeamSearchScorer' from 'transformers' (/usr/local/lib/python3.10/dist-packages/transformers/init.py)