Spaces:
Running
Running
File size: 4,954 Bytes
72e17aa | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 | import gradio as gr
from groq import Groq
import os
from whisper_support import transcribe
from kokoro_support import generate_tts
groq_api_key = os.environ.get("GROQ_API_KEY")
client = Groq(api_key = groq_api_key)
model_list =[
'openai/gpt-oss-120b',
'moonshotai/kimi-k2-instruct',
'meta-llama/llama-4-scout-17b-16e-instruct',
'openai/gpt-oss-20b',
'qwen/qwen3-32b',
'llama-3.3-70b-versatile',
'moonshotai/kimi-k2-instruct-0905',
'allam-2-7b',
'meta-llama/llama-4-maverick-17b-128e-instruct',
'llama-3.1-8b-instant',
]
reasoning_models = [
'openai/gpt-oss-120b',
'openai/gpt-oss-20b',
'qwen/qwen3-32b',
]
default_sys_prompt = """You are a helpful chatbot. You help the end user to the best of your ability."""
chat_history = []
def groq_voicechat(new_message: tuple, chat_history: list[dict], model: str, system_prompt: str, ):
'''
Groq chat API call wrapper.
inputs:
- model [str]: model from model_list (cbf static typing from the list)
- chat_history [list[dict]]: list of dictionaries of chat hist, needs "role" and "content" vars as strings
- new_message [tuple]: new user input message (assuming we're only accepting user inputs) from voice recording, to be transcribed.
- system prompt [str]: optional system prompt for whatever chat you're using
outputs:
- "" - used to delete old input msg in chat textbox lol
- nonsys_msg_hist [list[dict]]: updated chat history
'''
if model not in model_list:
raise ValueError(f"model must be in model_list: {model_list}")
return
#augment chat hist
nonsys_msg_hist = [{key: x[key] for key in ["role", "content"] if key in x} for x in chat_history] #clean the chatbot bullshit out
print(nonsys_msg_hist)
text_input = transcribe(new_message)
if transcription:
if transcription.startswith("Error"):
transcription = "Error in audio transcription."
return "error lol idk make this better later"
nonsys_msg_hist.extend(
[
{
"role": "user",
"content": text_input,
}
]
)
# use sys prompt
input_msg_hist = [
{
"role": "system",
"content": system_prompt,
}
]
input_msg_hist.extend(nonsys_msg_hist)
if model in reasoning_models:
chat_completion = client.chat.completions.create(
messages = input_msg_hist,
model = model,
include_reasoning = False, #removes reasoning tokens from output because I'm lazy
)
else:
chat_completion = client.chat.completions.create(
messages = input_msg_hist,
model = model,
# include_reasoning = False, #removes reasoning tokens from output because I'm lazy
)
output_msg = chat_completion.choices[0].message.content
# add to chat hist
nonsys_msg_hist.extend(
[
{
"role": "assistant",
"content": output_msg
}
]
)
return nonsys_msg_hist
def create_demo():
with gr.Blocks() as demo:
with gr.Row():
model = gr.Dropdown(model_list,
)
with gr.Row():
system_prompt = gr.Textbox(
value=default_sys_prompt,
interactive=True
)
with gr.Row():
chatbot = gr.Chatbot(label="Conversation", type="messages")
with gr.Row():
voiceinput = gr.Audio(
label="Input Audio",
sources=["microphone"],
type="numpy",
streaming=False,
)
with gr.Row():
clear = gr.ClearButton([voiceinput, chatbot], variant = 'stop')
with gr.Row():
playback_button = gr.Button("playback last message")
with gr.Row():
audio_out = gr.Audio(
label = "Output Audio",
interactive = False,
autoplay = True
)
voiceinput.stop_recording(
groq_voicechat,
[voiceinput, chatbot, model, system_prompt, ],
[chatbot]
) #WHAT AM I DOING LOL - COME BACK TO THIS
def playback_last_message(chat_history):
if len(chat_history) > 0:
last_message = chat_history[-1]['content']
return generate_tts(last_message)
return None
playback_button.click(
playback_last_message,
inputs=[chatbot],
outputs=[audio_out]
)
return demo
if __name__ == "__main__":
demo = create_demo()
demo.launch(
auth=("DigitalChild", "IhateBroccoli123"),
ssr_mode=False,
share=True,
)
|