fine-tuning error...

(this is my first experience in TTS)
My goal is to do fine-tuning on the model "tts_models--multilingual--multi-dataset--your_tts"
I use a dataset of 200 clips (22050)
I always get a message about "audio_unique names"...

....

TRAINING (2023-06-27 10:03:16)
! Run is kept in /home/ros/yourTTS/louisOutput1/vits_tts-louis-french-June-27-2023_10+03AM-d6581942
Traceback (most recent call last):
File "/home/ros/.local/lib/python3.10/site-packages/trainer/trainer.py", line 1591, in fit
self._fit()
File "/home/ros/.local/lib/python3.10/site-packages/trainer/trainer.py", line 1544, in _fit
self.train_epoch()
File "/home/ros/.local/lib/python3.10/site-packages/trainer/trainer.py", line 1309, in train_epoch
_, _ = self.train_step(batch, batch_num_steps, cur_step, loader_start_time)
File "/home/ros/.local/lib/python3.10/site-packages/trainer/trainer.py", line 1126, in train_step
batch = self.format_batch(batch)
File "/home/ros/.local/lib/python3.10/site-packages/trainer/trainer.py", line 912, in format_batch
batch = self.model.format_batch(batch)
File "/home/ros/TTS/TTS/tts/models/vits.py", line 1480, in format_batch
d_vectors = [d_vector_mapping[w]["embedding"] for w in batch["audio_unique_names"]]
File "/home/ros/TTS/TTS/tts/models/vits.py", line 1480, in
d_vectors = [d_vector_mapping[w]["embedding"] for w in batch["audio_unique_names"]]
KeyError: 'lr#wavs/a0134'

I am stuck...
Can anybody help me?....
THANKS!