import gradio as gr import json import torch import time import random from tqdm import tqdm from huggingface_hub import snapshot_download from models import AudioDiffusion, DDPMScheduler from audioldm.audio.stft import TacotronSTFT from audioldm.variational_autoencoder import AutoencoderKL from pydub import AudioSegment max_64_bit_int = 2**63 - 1 # Automatic device detection if torch.cuda.is_available(): device_type = "cuda" device_selection = "cuda:0" else: device_type = "cpu" device_selection = "cpu" class Tango: def __init__(self, name = "declare-lab/tango2", device = device_selection): path = snapshot_download(repo_id = name) vae_config = json.load(open("{}/vae_config.json".format(path))) stft_config = json.load(open("{}/stft_config.json".format(path))) main_config = json.load(open("{}/main_config.json".format(path))) self.vae = AutoencoderKL(**vae_config).to(device) self.stft = TacotronSTFT(**stft_config).to(device) self.model = AudioDiffusion(**main_config).to(device) vae_weights = torch.load("{}/pytorch_model_vae.bin".format(path), map_location = device) stft_weights = torch.load("{}/pytorch_model_stft.bin".format(path), map_location = device) main_weights = torch.load("{}/pytorch_model_main.bin".format(path), map_location = device) self.vae.load_state_dict(vae_weights) self.stft.load_state_dict(stft_weights) self.model.load_state_dict(main_weights) print ("Successfully loaded checkpoint from:", name) self.vae.eval() self.stft.eval() self.model.eval() self.scheduler = DDPMScheduler.from_pretrained(main_config["scheduler_name"], subfolder = "scheduler") def chunks(self, lst, n): # Yield successive n-sized chunks from a list for i in range(0, len(lst), n): yield lst[i:i + n] def generate(self, prompt, steps = 100, guidance = 3, samples = 1, disable_progress = True): # Generate audio for a single prompt string with torch.no_grad(): latents = self.model.inference([prompt], self.scheduler, steps, guidance, samples, disable_progress = disable_progress) mel = self.vae.decode_first_stage(latents) wave = self.vae.decode_to_waveform(mel) return wave def generate_for_batch(self, prompts, steps = 200, guidance = 3, samples = 1, batch_size = 8, disable_progress = True): # Generate audio for a list of prompt strings outputs = [] for k in tqdm(range(0, len(prompts), batch_size)): batch = prompts[k: k + batch_size] with torch.no_grad(): latents = self.model.inference(batch, self.scheduler, steps, guidance, samples, disable_progress = disable_progress) mel = self.vae.decode_first_stage(latents) wave = self.vae.decode_to_waveform(mel) outputs += [item for item in wave] if samples == 1: return outputs return list(self.chunks(outputs, samples)) # Initialize TANGO tango = Tango(device = "cpu") tango.vae.to(device_type) tango.stft.to(device_type) tango.model.to(device_type) def update_seed(is_randomize_seed, seed): if is_randomize_seed: return random.randint(0, max_64_bit_int) return seed def check( prompt, output_number, steps, guidance, is_randomize_seed, seed ): if prompt is None or prompt == "": raise gr.Error("Please provide a prompt input.") if not output_number in [1, 2, 3]: raise gr.Error("Please ask for 1, 2 or 3 output files.") def update_output(output_format, output_number): return [ gr.update(format = output_format), gr.update(format = output_format, visible = (2 <= output_number)), gr.update(format = output_format, visible = (output_number == 3)), gr.update(visible = False) ] def text2audio( prompt, output_number, steps, guidance, is_randomize_seed, seed ): start = time.time() if seed is None: seed = random.randint(0, max_64_bit_int) random.seed(seed) torch.manual_seed(seed) output_wave = tango.generate(prompt, steps, guidance, output_number) output_wave_1 = gr.make_waveform((16000, output_wave[0])) output_wave_2 = gr.make_waveform((16000, output_wave[1])) if (2 <= output_number) else None output_wave_3 = gr.make_waveform((16000, output_wave[2])) if (output_number == 3) else None end = time.time() secondes = int(end - start) minutes = secondes // 60 secondes = secondes - (minutes * 60) hours = minutes // 60 minutes = minutes - (hours * 60) return [ output_wave_1, output_wave_2, output_wave_3, gr.update(visible = True, value = "Start again to get a different result. The output have been generated in " + ((str(hours) + " h, ") if hours != 0 else "") + ((str(minutes) + " min, ") if hours != 0 or minutes != 0 else "") + str(secondes) + " sec.") ] # Gradio interface with gr.Blocks() as interface: gr.Markdown("""
Text-to-Audio
Generates 10 seconds of sound effects from description, freely, without account, without watermark.