{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Text-to-Speech LightSpeech"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"LightSpeech, Text to Melspectrogram."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"\n",
"This tutorial is available as an IPython notebook at [malaya-speech/example/tts-lightspeech](https://github.com/huseinzol05/malaya-speech/tree/master/example/tts-lightspeech).\n",
" \n",
"
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"\n",
"This module is not language independent, so it not save to use on different languages. Pretrained models trained on hyperlocal languages.\n",
" \n",
"
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"\n",
"This is an application of malaya-speech Pipeline, read more about malaya-speech Pipeline at [malaya-speech/example/pipeline](https://github.com/huseinzol05/malaya-speech/tree/master/example/pipeline).\n",
" \n",
"
"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import malaya_speech\n",
"import numpy as np\n",
"from malaya_speech import Pipeline\n",
"import matplotlib.pyplot as plt\n",
"import IPython.display as ipd"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### LightSpeech description\n",
"\n",
"1. Malaya-speech FastSpeech2 will generate melspectrogram with feature size 80.\n",
"2. Use Malaya-speech vocoder to convert melspectrogram to waveform.\n",
"3. Cannot generate more than melspectrogram longer than 2000 timestamp, it will throw an error. Make sure the texts are not too long."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### List available LightSpeech"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"scrolled": true
},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" Size (MB) \n",
" Quantized Size (MB) \n",
" Understand punctuation \n",
" Is lowercase \n",
" \n",
" \n",
" \n",
" \n",
" osman \n",
" 167 \n",
" 43.3 \n",
" True \n",
" False \n",
" \n",
" \n",
" yasmin \n",
" 167 \n",
" 43.3 \n",
" True \n",
" False \n",
" \n",
" \n",
"
\n",
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"fig = plt.figure(figsize=(10, 8))\n",
"ax1 = fig.add_subplot(311)\n",
"ax1.set_title('Predicted Mel-Spectrogram')\n",
"im = ax1.imshow(np.rot90(r_male['mel-output']), aspect='auto', interpolation='none')\n",
"fig.colorbar(mappable=im, shrink=0.65, orientation='horizontal', ax=ax1)\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Load Vocoder model\n",
"\n",
"There are 2 ways to synthesize melspectrogram output from TTS models,\n",
"\n",
"1. If you are going to use individual speaker vocoder, make sure the speakers are the same If use female tacotron2, need to use female MelGAN also. Use `mel-output` from TTS model. Read more at https://malaya-speech.readthedocs.io/en/latest/load-vocoder.html\n",
"2. If you are going to use universal MelGAN, use `universal-output` from TTS model. Read more at https://malaya-speech.readthedocs.io/en/latest/load-universal-melgan.html\n",
"\n",
"We prefer individual speaker vocoder, size only 17MB and faster than universal vocoder."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "7276552c6d3a46d4b05f108f07d472f5",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='Downloading', max=78672409.0, style=ProgressStyle(descrip…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
}
],
"source": [
"universal_melgan = malaya_speech.vocoder.melgan(model = 'universal-1024')"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"y_ = universal_melgan(r_male['universal-output'])\n",
"ipd.Audio(y_, rate = 22050)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Playing around with speed, f0 and energy ratio"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
"r_male = male.predict(string1, speed_ratio = 1.5)"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"y_ = universal_melgan(r_male['universal-output'])\n",
"ipd.Audio(y_, rate = 22050)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"r_male = male.predict(string1, f0_ratio = -1.5)"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"y_ = universal_melgan(r_male['universal-output'])\n",
"ipd.Audio(y_, rate = 22050)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Combined everything using Pipeline"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"p = Pipeline()\n",
"pipeline = (\n",
" p.map(male)\n",
" .map(lambda x: x['universal-output'])\n",
" .map(universal_melgan)\n",
")\n",
"p.visualize()"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
"r = p('husein wangi tetapi ketiak masam nasib baik kacak')"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"ipd.Audio(r['vocoder-melgan'], rate = 22050)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Compare speed with Tacotron2"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
"# https://www.hmetro.com.my/mutakhir/2020/12/657604/6-cadangan-tangani-kelemahan-kawal-selia-halal\n",
"\n",
"text = 'Kuasa pensijilan halal, dan penguatkuasaan halal terletak di bawah bidang kuasa agensi yang berbeza.'"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d5a20fc50bca4f668649f04f53b489ad",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"HBox(children=(FloatProgress(value=0.0, description='Downloading', max=104374569.0, style=ProgressStyle(descri…"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
}
],
"source": [
"male_tacotron2 = malaya_speech.tts.tacotron2(model = 'osman')"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"CPU times: user 2.9 s, sys: 59.7 ms, total: 2.96 s\n",
"Wall time: 2.66 s\n"
]
}
],
"source": [
"%%time\n",
"\n",
"r_male_tacotron2 = male_tacotron2.predict(text)"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"y_ = universal_melgan(r_male_tacotron2['universal-output'])\n",
"ipd.Audio(y_, rate = 22050)"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"CPU times: user 235 ms, sys: 48.5 ms, total: 284 ms\n",
"Wall time: 61.6 ms\n"
]
}
],
"source": [
"%%time\n",
"\n",
"r_male = male.predict(text)"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 21,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"y_ = universal_melgan(r_male['universal-output'])\n",
"ipd.Audio(y_, rate = 22050)"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"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": "3.8.10"
},
"varInspector": {
"cols": {
"lenName": 16,
"lenType": 16,
"lenVar": 40
},
"kernels_config": {
"python": {
"delete_cmd_postfix": "",
"delete_cmd_prefix": "del ",
"library": "var_list.py",
"varRefreshCmd": "print(var_dic_list())"
},
"r": {
"delete_cmd_postfix": ") ",
"delete_cmd_prefix": "rm(",
"library": "var_list.r",
"varRefreshCmd": "cat(var_dic_list()) "
}
},
"types_to_exclude": [
"module",
"function",
"builtin_function_or_method",
"instance",
"_Feature"
],
"window_display": false
}
},
"nbformat": 4,
"nbformat_minor": 4
}