fix bug

examples/industrial_data_pretraining/llm_asr/demo_speech2text.py

@@ -16,12 +16,14 @@ jsonl = (
 with open(jsonl, "r") as f:
     lines = f.readlines()
 
+tearchforing = True
 for i, line in enumerate(lines):
     data_dict = json.loads(line.strip())
     data = data_dict["messages"]
 
     res = model.generate(
-        input=data,
+        input=[data],
+        tearchforing=tearchforing,
         cache={},
     )
 

funasr/models/llm_asr/model.py

@@ -568,6 +568,7 @@ class LLMASR2(nn.Module):
             [],
             [],
             [],
+            [],
         )
 
         for i, (system_prompt, user_prompt, target_out) in enumerate(zip(system, user, assistant)):
@@ -624,7 +625,7 @@ class LLMASR2(nn.Module):
         input_ids = torch.tensor(input_ids, dtype=torch.int64)  # [: self.max_token_length]
         attention_mask = torch.tensor([1] * len(input_ids), dtype=torch.int32)
         labels = torch.tensor(labels, dtype=torch.int64)  # [: self.max_token_length]
-        source_ids = torch.tensor(source_ids, dtype=torch.int64)
+        source_ids = torch.tensor(source_ids_i, dtype=torch.int64)
         target_ids = torch.tensor(target_ids, dtype=torch.int64)
 
         fbank = speech[0, :, :]
@@ -662,7 +663,7 @@ class LLMASR2(nn.Module):
         if kwargs.get("batch_size", 1) > 1:
             raise NotImplementedError("batch decoding is not implemented")
 
-        contents = self.data_template(data_in)
+        contents = self.data_template(data_in[0])
         output = self.data_load_speech(contents, tokenizer, frontend, **kwargs)
         batch = to_device(output, kwargs["device"])
 
@@ -676,7 +677,7 @@ class LLMASR2(nn.Module):
 
         input_ids = batch["input_ids"]
         source_ids = batch["source_ids"]
-        if kwargs.get("tearchforing", False):
+        if not kwargs.get("tearchforing", False):
             input_ids = source_ids
         input_ids[input_ids < 0] = 0
         inputs_embeds = self.llm.model.get_input_embeddings()(input_ids)
@@ -704,6 +705,23 @@ class LLMASR2(nn.Module):
                 generated_ids, skip_special_tokens=kwargs.get("skip_special_tokens", True)
             )[0]
             label = contents["assistant"][0]
+            loss = None
+        else:
+
+            labels_ids = batch["labels_ids"]
+            labels_ids[labels_ids == -1] = -100
+            attention_mask = batch.get("attention_mask", None)
+            model_outputs = self.llm(
+                inputs_embeds=inputs_embeds, attention_mask=attention_mask, labels=labels_ids
+            )
+
+            preds = torch.argmax(model_outputs.logits, -1)[:, source_ids.shape[1]]
+            response = tokenizer.batch_decode(
+                preds,
+                add_special_tokens=False,
+                skip_special_tokens=kwargs.get("skip_special_tokens", True),
+            )[0]
+            loss = model_outputs.loss
 
         ibest_writer = None
         if kwargs.get("output_dir") is not None:
@@ -713,10 +731,12 @@ class LLMASR2(nn.Module):
 
         results = []
         result_i = {"key": key[0], "text": response, "label": label}
+        if loss is not None:
+            result_i["loss"] = loss
         results.append(result_i)
 
         if ibest_writer is not None:
-            ibest_writer["text"][key[0]] = text
+            ibest_writer["text"][key[0]] = response
             ibest_writer["label"][key[0]] = label
 
         return results, meta_data