Upload processor

chat_template.jinja

@@ -1,9 +1,9 @@
 {%- set system_prompt = system_prompt | default(" Transform the text provided by various speakers into speech output, utilizing the distinct voice of each respective speaker.
 ") -%}
 {{ system_prompt -}}
-{%- set speech_start_token = speech_start_token | default("<|vision_start|>") %}
-{%- set speech_end_token = speech_end_token | default("<|vision_end|>") %}
-{%- set speech_diffusion_token = speech_diffusion_token | default("<|vision_pad|>") %}
+{%- set audio_bos_token = audio_bos_token | default("<|vision_start|>") %}
+{%- set audio_eos_token = audio_eos_token | default("<|vision_end|>") %}
+{%- set audio_diffusion_token = audio_diffusion_token | default("<|vision_pad|>") %}
 {%- set ns = namespace(speakers_with_audio="") %}
 {%- for message in messages %}
     {%- set role = message['role'] %}
@@ -19,7 +19,7 @@
 " }}
 {%- for speaker in ns.speakers_with_audio.rstrip(',').split(',') %}
 {%- if speaker %}
- Speaker {{ speaker }}:{{ speech_start_token }}{{ speech_diffusion_token }}{{ speech_end_token }}{{ "
+ Speaker {{ speaker }}:{{ audio_bos_token }}{{ audio_diffusion_token }}{{ audio_eos_token }}{{ "
 " }}
 {%- endif %}
 {%- endfor %}
@@ -36,4 +36,4 @@
     {%- endfor %}
 {%- endfor %}
  Speech output:{{ "
-" }}{{ speech_start_token }}
+" }}{{ audio_bos_token }}