You have to change the code a little as well. You have to comment out where speaker_id is being referenced in the code. I commented out 3 lines and it works fine for me without further modificaiton;

[gangagyatso4364]

          You have to change the code a little as well. You have to comment out where speaker_id is being referenced in the code. I commented out 3 lines and it works fine for me without further modificaiton;

1-
model_outputs = model(
input_ids=batch["input_ids"],
attention_mask=batch["attention_mask"],
labels=batch["labels"],
labels_attention_mask=batch["labels_attention_mask"],
# speaker_id=batch["speaker_id"],
return_dict=True,
monotonic_alignment_function=maximum_path,
)

2-model_outputs_train = model(
input_ids=batch["input_ids"],
attention_mask=batch["attention_mask"],
labels=batch["labels"],
labels_attention_mask=batch["labels_attention_mask"],
# speaker_id=batch["speaker_id"],
return_dict=True,
monotonic_alignment_function=maximum_path,
)

3- model_outputs_train = model(
input_ids=batch["input_ids"],
attention_mask=batch["attention_mask"],
labels=batch["labels"],
labels_attention_mask=batch["labels_attention_mask"],
# speaker_id=batch["speaker_id"],
return_dict=True,
monotonic_alignment_function=maximum_path,
)

Originally posted by @muhammadsaadgondal in #35 (comment)

[gangagyatso4364]

Can you share the config file training multiple speakers. ?

[C0deXG]

Can you share the config file training multiple speakers. ?

Just remove all the speaker_id or speaker mentioing in config file so remove all speaker mentioning in the config