import os
import requests
import torch
import scipy.io.wavfile as wav
import streamlit as st
from io import BytesIO
from transformers import (
AutoTokenizer,
AutoModelForCausalLM,
pipeline,
AutoProcessor,
MusicgenForConditionalGeneration
)
from streamlit_lottie import st_lottie
# ---------------------------------------------------------------------
# 1) PAGE CONFIGURATION
# ---------------------------------------------------------------------
st.set_page_config(
page_title="AI Radio Imaging with Llama 3",
page_icon="🎧",
layout="wide"
)
# ---------------------------------------------------------------------
# 2) CUSTOM CSS / UI DESIGN
# ---------------------------------------------------------------------
CUSTOM_CSS = """
"""
st.markdown(CUSTOM_CSS, unsafe_allow_html=True)
# ---------------------------------------------------------------------
# 3) LOAD LOTTIE ANIMATION
# ---------------------------------------------------------------------
@st.cache_data
def load_lottie_url(url: str):
r = requests.get(url)
if r.status_code != 200:
return None
return r.json()
LOTTIE_URL = "https://assets3.lottiefiles.com/temp/lf20_Q6h5zV.json"
lottie_animation = load_lottie_url(LOTTIE_URL)
# ---------------------------------------------------------------------
# 4) LOAD LLAMA 3 (GATED MODEL)
# ---------------------------------------------------------------------
@st.cache_resource
def load_llama_pipeline(model_id: str, device: str, token: str):
tokenizer = AutoTokenizer.from_pretrained(model_id, use_auth_token=token)
model = AutoModelForCausalLM.from_pretrained(
model_id,
use_auth_token=token,
torch_dtype=torch.float16 if device == "auto" else torch.float32,
device_map=device,
low_cpu_mem_usage=True
)
text_gen_pipeline = pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
device_map=device
)
return text_gen_pipeline
# ---------------------------------------------------------------------
# 5) GENERATE RADIO SCRIPT
# ---------------------------------------------------------------------
def generate_radio_script(user_input: str, pipeline_llama) -> str:
system_prompt = (
"You are a top-tier radio imaging producer using Llama 3. "
"Take the user's concept and craft a short, creative promo script."
)
combined_prompt = f"{system_prompt}\nUser concept: {user_input}\nRefined script:"
result = pipeline_llama(
combined_prompt,
max_new_tokens=200,
do_sample=True,
temperature=0.9
)
output_text = result[0]["generated_text"]
if "Refined script:" in output_text:
output_text = output_text.split("Refined script:", 1)[-1].strip()
output_text += "\n\n(Generated by Llama 3 - Radio Imaging)"
return output_text
# ---------------------------------------------------------------------
# 6) LOAD MUSICGEN
# ---------------------------------------------------------------------
@st.cache_resource
def load_musicgen_model():
mg_model = MusicgenForConditionalGeneration.from_pretrained("facebook/musicgen-small")
mg_processor = AutoProcessor.from_pretrained("facebook/musicgen-small")
return mg_model, mg_processor
# ---------------------------------------------------------------------
# 7) HEADER
# ---------------------------------------------------------------------
st.title("🎧 AI Radio Imaging with Llama 3")
st.subheader("Create engaging radio promos with Llama 3 + MusicGen")
st.markdown("""Create **radio imaging promos** and **jingles** easily. Ensure you have access to
**meta-llama/Meta-Llama-3-70B** on Hugging Face and provide your token below.""")
if lottie_animation:
st_lottie(lottie_animation, height=180, loop=True, key="radio_lottie")
st.markdown("---")
# ---------------------------------------------------------------------
# 8) USER INPUT
# ---------------------------------------------------------------------
st.subheader("🎤 Step 1: Describe Your Promo Idea")
prompt = st.text_area(
"Example: 'A 15-second hype jingle for a morning talk show, fun and energetic.'",
height=120
)
col_model, col_device = st.columns(2)
with col_model:
llama_model_id = st.text_input(
"Llama 3 Model ID",
value="meta-llama/Meta-Llama-3-70B",
help="Enter the exact model ID from Hugging Face."
)
with col_device:
device_option = st.selectbox(
"Device",
["auto", "cpu"],
help="Choose GPU (auto) or CPU."
)
hf_token = os.getenv("HF_TOKEN")
if not hf_token:
st.error("No HF_TOKEN found. Please set it in your environment.")
st.stop()
if st.button("✍ Generate Promo Script"):
if not prompt.strip():
st.error("Please provide a concept first.")
else:
with st.spinner("Generating script..."):
try:
llama_pipeline = load_llama_pipeline(llama_model_id, device_option, hf_token)
final_script = generate_radio_script(prompt, llama_pipeline)
st.success("Promo script generated!")
st.text_area("Generated Script", value=final_script, height=200)
except Exception as e:
st.error(f"Llama generation error: {e}")
st.markdown("---")
# ---------------------------------------------------------------------
# 9) GENERATE AUDIO WITH MUSICGEN
# ---------------------------------------------------------------------
st.subheader("🎵 Step 2: Generate Audio")
audio_length = st.slider("Track Length (tokens)", 128, 1024, 512, 64)
if st.button("🎧 Create Audio"):
if "final_script" not in st.session_state:
st.error("Please generate a script first.")
else:
with st.spinner("Generating audio..."):
try:
mg_model, mg_processor = load_musicgen_model()
inputs = mg_processor(
text=[st.session_state["final_script"]],
padding=True,
return_tensors="pt"
)
audio_values = mg_model.generate(**inputs, max_new_tokens=audio_length)
sr = mg_model.config.audio_encoder.sampling_rate
output_file = "radio_jingle.wav"
audio_data = audio_values[0, 0].cpu().numpy()
normalized_audio = (audio_data / max(abs(audio_data)) * 32767).astype("int16")
wav.write(output_file, rate=sr, data=normalized_audio)
st.success("Audio generated! Play it below:")
st.audio(output_file)
except Exception as e:
st.error(f"MusicGen error: {e}")
# ---------------------------------------------------------------------
# 10) FOOTER
# ---------------------------------------------------------------------
st.markdown("---")
st.markdown(
"""
""",
unsafe_allow_html=True
)