mirror of https://github.com/coqui-ai/TTS.git
381 lines
10 KiB
Plaintext
381 lines
10 KiB
Plaintext
{
|
||
"cells": [
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"%load_ext autoreload\n",
|
||
"%autoreload 2\n",
|
||
"import os\n",
|
||
"import sys\n",
|
||
"import io\n",
|
||
"import torch \n",
|
||
"import time\n",
|
||
"import numpy as np\n",
|
||
"from collections import OrderedDict\n",
|
||
"from matplotlib import pylab as plt\n",
|
||
"\n",
|
||
"%pylab inline\n",
|
||
"rcParams[\"figure.figsize\"] = (16,5)\n",
|
||
"sys.path.append('/home/erogol/projects/')\n",
|
||
"\n",
|
||
"import librosa\n",
|
||
"import librosa.display\n",
|
||
"\n",
|
||
"from TTS.models.tacotron import Tacotron \n",
|
||
"from TTS.layers import *\n",
|
||
"from TTS.utils.data import *\n",
|
||
"from TTS.utils.audio import AudioProcessor\n",
|
||
"from TTS.utils.generic_utils import load_config\n",
|
||
"from TTS.utils.text import text_to_sequence\n",
|
||
"\n",
|
||
"import IPython\n",
|
||
"from IPython.display import Audio\n",
|
||
"from utils import *"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"def tts(model, text, CONFIG, use_cuda, ap, figures=True):\n",
|
||
" t_1 = time.time()\n",
|
||
" waveform, alignment, spectrogram, stop_tokens = create_speech(model, text, CONFIG, use_cuda, ap) \n",
|
||
" print(\" > Run-time: {}\".format(time.time() - t_1))\n",
|
||
" if figures: \n",
|
||
" visualize(alignment, spectrogram, stop_tokens, CONFIG) \n",
|
||
" IPython.display.display(Audio(waveform, rate=CONFIG.sample_rate)) \n",
|
||
" out_path = 'benchmark_samples/'\n",
|
||
" os.makedirs(out_path, exist_ok=True)\n",
|
||
" file_name = text.replace(\" \", \"_\").replace(\".\",\"\") + \".wav\"\n",
|
||
" out_path = os.path.join(out_path, file_name)\n",
|
||
" ap.save_wav(waveform, out_path)\n",
|
||
" return alignment, spectrogram, stop_tokens"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"# Set constants\n",
|
||
"ROOT_PATH = '/data/shared/erogol_models/May-22-2018_03:24PM-loc-sen-attn-e6112f7/'\n",
|
||
"MODEL_PATH = ROOT_PATH + '/checkpoint_272976.pth.tar'\n",
|
||
"CONFIG_PATH = ROOT_PATH + '/config.json'\n",
|
||
"OUT_FOLDER = ROOT_PATH + '/test/'\n",
|
||
"CONFIG = load_config(CONFIG_PATH)\n",
|
||
"use_cuda = True"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"# load the model\n",
|
||
"model = Tacotron(CONFIG.embedding_size, CONFIG.num_freq, CONFIG.num_mels, CONFIG.r)\n",
|
||
"\n",
|
||
"# load the audio processor\n",
|
||
"\n",
|
||
"ap = AudioProcessor(CONFIG.sample_rate, CONFIG.num_mels, CONFIG.min_level_db,\n",
|
||
" CONFIG.frame_shift_ms, CONFIG.frame_length_ms, CONFIG.preemphasis,\n",
|
||
" CONFIG.ref_level_db, CONFIG.num_freq, CONFIG.power, griffin_lim_iters=30) \n",
|
||
"\n",
|
||
"\n",
|
||
"# load model state\n",
|
||
"if use_cuda:\n",
|
||
" cp = torch.load(MODEL_PATH)\n",
|
||
"else:\n",
|
||
" cp = torch.load(MODEL_PATH, map_location=lambda storage, loc: storage)\n",
|
||
"\n",
|
||
"# load the model\n",
|
||
"model.load_state_dict(cp['model'])\n",
|
||
"if use_cuda:\n",
|
||
" model.cuda()\n",
|
||
"model.eval()"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"metadata": {},
|
||
"source": [
|
||
"### EXAMPLES FROM TRAINING SET"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"import pandas as pd\n",
|
||
"df = pd.read_csv('/data/shared/KeithIto/LJSpeech-1.0/metadata_val.csv', delimiter='|')"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {
|
||
"scrolled": false
|
||
},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = df.iloc[175, 1]\n",
|
||
"print(sentence)\n",
|
||
"model.decoder.max_decoder_steps = 250\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"metadata": {},
|
||
"source": [
|
||
"### Comparision with https://mycroft.ai/blog/available-voices/"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {
|
||
"scrolled": false
|
||
},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"It took me quite a long time to develop a voice, and now that I have it I'm not going to be silent.\"\n",
|
||
"model.decoder.max_decoder_steps = 250\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap, figures=True)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"Be a voice,not an echo.\" # 'echo' is not in training set. \n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"The human voice is the most perfect instrument of all.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"I'm sorry Dave. I'm afraid I can't do that.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"This cake is great. It's so delicious and moist.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"metadata": {},
|
||
"source": [
|
||
"### Comparison with https://keithito.github.io/audio-samples/"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"Generative adversarial network or variational auto-encoder.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"Scientists at the CERN laboratory say they have discovered a new particle.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"Here’s a way to measure the acute emotional intelligence that has never gone out of style.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"President Trump met with other leaders at the Group of 20 conference.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"The buses aren't the problem, they actually provide a solution.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"metadata": {},
|
||
"source": [
|
||
"### Comparison with https://google.github.io/tacotron/publications/tacotron/index.html"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"Generative adversarial network or variational auto-encoder.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"Basilar membrane and otolaryngology are not auto-correlations.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"He has read the whole thing.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"He reads books.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"Thisss isrealy awhsome.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"This is your internet browser, Firefox.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"This is your internet browser Firefox.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"The quick brown fox jumps over the lazy dog.\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"sentence = \"Does the quick brown fox jump over the lazy dog?\"\n",
|
||
"align, spec, stop_tokens = tts(model, sentence, CONFIG, use_cuda, ap)"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": null,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"!zip benchmark_samples/samples.zip benchmark_samples/*"
|
||
]
|
||
}
|
||
],
|
||
"metadata": {
|
||
"kernelspec": {
|
||
"display_name": "Python 3",
|
||
"language": "python",
|
||
"name": "python3"
|
||
},
|
||
"language_info": {
|
||
"codemirror_mode": {
|
||
"name": "ipython",
|
||
"version": 3
|
||
},
|
||
"file_extension": ".py",
|
||
"mimetype": "text/x-python",
|
||
"name": "python",
|
||
"nbconvert_exporter": "python",
|
||
"pygments_lexer": "ipython3",
|
||
"version": "3.6.5"
|
||
}
|
||
},
|
||
"nbformat": 4,
|
||
"nbformat_minor": 2
|
||
}
|