Upload processor

chat_template.jinja

@@ -15,23 +15,18 @@
 {{- content -}}
 {%- endif -%}
 {%- endmacro -%}
-
 {%- for m in messages -%}
 {%- if m.role == 'system' -%}
 <|system|>
 {{ to_text(m.content) | trim }}
-
 {%- elif m.role == 'user' -%}
 <|user|>
 {{ to_text(m.content) | trim }}
-
 {%- elif m.role == 'assistant' -%}
 <|assistant|>
 {{ to_text(m.content) | trim }}
-
 {%- endif -%}
 {%- endfor -%}
-
 {%- if add_generation_prompt -%}
 <|assistant|>
 {% endif -%}

processor_config.json

@@ -1,5 +1,6 @@
 {
   "audio_token": "<|pad|>",
+  "default_transcription_prompt": "Please transcribe this audio into text",
   "feature_extractor": {
     "chunk_length": 30,
     "dither": 0.0,
@@ -14,5 +15,6 @@
     "return_attention_mask": false,
     "sampling_rate": 16000
   },
+  "max_audio_len": 600,
   "processor_class": "GlmAsrProcessor"
 }

tokenizer_config.json

@@ -15,7 +15,12 @@
     "<|assistant|>",
     "<|observation|>",
     "<|begin_of_image|>",
-    "<|end_of_image|>"
+    "<|end_of_image|>",
+    "<|begin_of_video|>",
+    "<|end_of_video|>",
+    "<|pad|>",
+    "<|begin_of_audio|>",
+    "<|end_of_audio|>"
   ],
   "is_local": false,
   "model_input_names": [