Spaces:
Running
on
Zero
Running
on
Zero
| import spaces | |
| import gradio as gr | |
| import json | |
| import torch | |
| import wavio | |
| from tqdm import tqdm | |
| from huggingface_hub import snapshot_download | |
| from pydub import AudioSegment | |
| from gradio import Markdown | |
| import uuid | |
| import torch | |
| from diffusers import DiffusionPipeline,AudioPipelineOutput | |
| from transformers import CLIPTextModel, T5EncoderModel, AutoModel, T5Tokenizer, T5TokenizerFast | |
| from typing import Union | |
| from diffusers.utils.torch_utils import randn_tensor | |
| from tqdm import tqdm | |
| from TangoFlux import TangoFluxInference | |
| import torchaudio | |
| # Define the description text | |
| description_text = """ | |
| # TangoFlux Text-to-Audio Generation | |
| Generate high-quality audio from text descriptions using TangoFlux. | |
| ## Instructions: | |
| 1. Enter your text description in the prompt box | |
| 2. Adjust the generation parameters if desired | |
| 3. Click submit to generate audio | |
| ## Parameters: | |
| - Steps: Higher values give better quality but take longer | |
| - Guidance Scale: Controls how closely the generation follows the prompt | |
| - Duration: Length of the generated audio in seconds | |
| """ | |
| tangoflux = TangoFluxInference(name="declare-lab/TangoFlux") | |
| def gradio_generate(prompt, steps, guidance, duration=10): | |
| output = tangoflux.generate(prompt, steps=steps, guidance_scale=guidance, duration=duration) | |
| filename = 'temp.wav' | |
| output = output[:,:int(duration*44100)] | |
| torchaudio.save(filename, output, 44100) | |
| return filename | |
| # Gradio input and output components | |
| input_text = gr.Textbox(lines=2, label="Prompt") | |
| output_audio = gr.Audio(label="Generated Audio", type="filepath") | |
| denoising_steps = gr.Slider(minimum=10, maximum=100, value=25, step=5, label="Steps", interactive=True) | |
| guidance_scale = gr.Slider(minimum=1, maximum=10, value=4.5, step=0.5, label="Guidance Scale", interactive=True) | |
| duration_scale = gr.Slider(minimum=1, maximum=30, value=10, step=1, label="Duration", interactive=True) | |
| # Gradio interface | |
| gr_interface = gr.Interface( | |
| fn=gradio_generate, | |
| inputs=[input_text, denoising_steps, guidance_scale, duration_scale], | |
| outputs=output_audio, | |
| title="TangoFlux: Super Fast and Faithful Text to Audio Generation with Flow Matching and Clap-Ranked Preference Optimization", | |
| description=description_text, | |
| allow_flagging=False, | |
| theme="Yntec/HaleyCH_Theme_Orange", | |
| examples=[ | |
| ["Quiet whispered conversation gradually fading into distant jet engine roar diminishing into silence"], | |
| ["Clear sound of bicycle tires crunching on loose gravel and dirt, followed by deep male laughter echoing"], | |
| ["Multiple ducks quacking loudly with splashing water and piercing wild animal shriek in background"], | |
| ["Powerful ocean waves crashing and receding on sandy beach with distant seagulls"], | |
| ["Gentle female voice cooing and baby responding with happy gurgles and giggles"], | |
| ["Clear male voice speaking, sharp popping sound, followed by genuine group laughter"], | |
| ["Stream of water hitting empty ceramic cup, pitch rising as cup fills up"], | |
| ["Massive crowd erupting in thunderous applause and excited cheering"], | |
| ["Deep rolling thunder with bright lightning strikes crackling through sky"], | |
| ["Aggressive dog barking and distressed cat meowing as racing car roars past at high speed"], | |
| ["Peaceful stream bubbling and birds singing, interrupted by sudden explosive gunshot"], | |
| ["Man speaking outdoors, goat bleating loudly, metal gate scraping closed, ducks quacking frantically, wind howling into microphone"], | |
| ["Series of loud aggressive dog barks echoing"], | |
| ["Multiple distinct cat meows at different pitches"], | |
| ["Rhythmic wooden table tapping overlaid with steady water pouring sound"], | |
| ["Sustained crowd applause with camera clicks and amplified male announcer voice"], | |
| ["Two sharp gunshots followed by panicked birds taking flight with rapid wing flaps"], | |
| ["Melodic human whistling harmonizing with natural birdsong"], | |
| ["Deep rhythmic snoring with clear breathing patterns"], | |
| ["Multiple racing engines revving and accelerating with sharp whistle piercing through"], | |
| ["Massive stadium crowd cheering as thunder crashes and lightning strikes"], | |
| ["Heavy helicopter blades chopping through air with engine and wind noise"], | |
| ["Dog barking excitedly and man shouting as race car engine roars past"], | |
| ["Generate the festive sounds of a fireworks show: explosions lighting up the sky, crowd cheering, and the faint music playing in the background!! Celebration of the new year!"], | |
| ["Melodic human whistling harmonizing with natural birdsong"], | |
| ["A parade marches through a town square, with drumbeats pounding, children clapping, and a horse neighing amidst the commotion"], | |
| ["Quiet speech and then and airplane flying away"], | |
| ["A soccer ball hits a goalpost with a metallic clang, followed by cheers, clapping, and the distant hum of a commentator's voice"], | |
| ["A basketball bounces rhythmically on a court, shoes squeak against the floor, and a referee's whistle cuts through the air"], | |
| ["Dripping water echoes sharply, a distant growl reverberates through the cavern, and soft scraping metal suggests something lurking unseen"], | |
| ["A cow is mooing whilst a lion is roaring in the background as a hunter shoots. A flock of birds subsequently fly away from the trees."], | |
| ["The deep growl of an alligator ripples through the swamp as reeds sway with a soft rustle and a turtle splashes into the murky water"], | |
| ["Gentle female voice cooing and baby responding with happy gurgles and giggles"], | |
| ['doorbell ding once followed by footsteps gradually getting louder and a door is opened '], | |
| ["A fork scrapes a plate, water drips slowly into a sink, and the faint hum of a refrigerator lingers in the background"], | |
| ["Powerful ocean waves crashing and receding on sandy beach with distant seagulls"], | |
| ["Emulate the lively sounds of a retro arcade: 8-bit game music, coins clinking. People cheering occasionally when players winning"], | |
| ["Simulate a forest ambiance with birds chirping and wind rustling through the leaves"], | |
| ["A train conductor blows a sharp whistle, metal wheels screech on the rails, and passengers murmur while settling into their seats"], | |
| ["Generate an energetic and bustling city street scene with distant traffic and close conversations"], | |
| ["Alarms blare with rising urgency as fragments clatter against a metallic hull, interrupted by a faint hiss of escaping air"], | |
| ["Create a serene soundscape of a quiet beach at sunset"], | |
| ["Tiny pops and hisses of chemical reactions intermingle with the rhythmic pumping of a centrifuge and the soft whirr of air filtration"], | |
| ["A train conductor blows a sharp whistle, metal wheels screech on the rails, and passengers murmur while settling into their seats"], | |
| ["Emulate the lively sounds of a retro arcade: 8-bit game music, coins clinking. People cheering occasionally when players winning"], | |
| ["Quiet whispered conversation gradually fading into distant jet engine roar diminishing into silence"], | |
| ["Clear sound of bicycle tires crunching on loose gravel and dirt, followed by deep male laughter echoing"], | |
| ["Multiple ducks quacking loudly with splashing water and piercing wild animal shriek in background"], | |
| ["Create the underwater soundscape: gentle waves, faint whale calls, and the occasional clink of scuba gear"], | |
| ["Recreate the sounds of an active volcano: rumbling earth, lava bubbling, and the occasional loud explosive roar of an eruption"], | |
| ["A pile of coins spills onto a wooden table with a metallic clatter, followed by the hushed murmur of a tavern crowd and the creak of a swinging door"], | |
| ["Clear male voice speaking, sharp popping sound, followed by genuine group laughter"], | |
| ["Stream of water hitting empty ceramic cup, pitch rising as cup fills up"], | |
| ["Massive crowd erupting in thunderous applause and excited cheering"], | |
| ["Deep rolling thunder with bright lightning strikes crackling through sky"], | |
| ["Aggressive dog barking and distressed cat meowing as racing car roars past at high speed"], | |
| ["Peaceful stream bubbling and birds singing, interrupted by sudden explosive gunshot"], | |
| ["Man speaking outdoors, goat bleating loudly, metal gate scraping closed, ducks quacking frantically, wind howling into microphone"], | |
| ["Series of loud aggressive dog barks echoing"], | |
| ["Multiple distinct cat meows at different pitches"], | |
| ["Rhythmic wooden table tapping overlaid with steady water pouring sound"], | |
| ["Sustained crowd applause with camera clicks and amplified male announcer voice"], | |
| ["Two sharp gunshots followed by panicked birds taking flight with rapid wing flaps"], | |
| ["Deep rhythmic snoring with clear breathing patterns"], | |
| ["Multiple racing engines revving and accelerating with sharp whistle piercing through"], | |
| ["Massive stadium crowd cheering as thunder crashes and lightning strikes"], | |
| ["Heavy helicopter blades chopping through air with engine and wind noise"], | |
| ["Dog barking excitedly and man shouting as race car engine roars past"], | |
| ["A bicycle peddling on dirt and gravel followed by a man speaking then laughing"], | |
| ["Ducks quack and water splashes with some animal screeching in the background"], | |
| ["Describe the sound of the ocean"], | |
| ["A woman and a baby are having a conversation"], | |
| ["A man speaks followed by a popping noise and laughter"], | |
| ["A cup is filled from a faucet"], | |
| ["An audience cheering and clapping"], | |
| ["Rolling thunder with lightning strikes"], | |
| ["A dog barking and a cat mewing and a racing car passes by"], | |
| ["Gentle water stream, birds chirping and sudden gun shot"], | |
| ["A dog barking"], | |
| ["A cat meowing"], | |
| ["Wooden table tapping sound while water pouring"], | |
| ["Applause from a crowd with distant clicking and a man speaking over a loudspeaker"], | |
| ["two gunshots followed by birds flying away while chirping"], | |
| ["Whistling with birds chirping"], | |
| ["A person snoring"], | |
| ["Motor vehicles are driving with loud engines and a person whistles"], | |
| ["People cheering in a stadium while thunder and lightning strikes"], | |
| ["A helicopter is in flight"], | |
| ["A dog barking and a man talking and a racing car passes by"], | |
| ], | |
| cache_examples="lazy", | |
| ) | |
| # Launch the interface | |
| gr_interface.queue(15).launch() |