initial commit

README.md

@@ -1,3 +1,153 @@
 ---
-license: apache-2.0
+language: "en"
+inference: false
+tags:
+- Vocoder
+- DiffWave
+- text-to-speech
+- TTS
+- speech-synthesis
+- speechbrain
+license: "apache-2.0"
+datasets:
+- LJSpeech
 ---
+
+# Vocoder with DiffWave trained on LJSpeech
+
+This repository provides all the necessary tools for using a [DiffWave](https://arxiv.org/pdf/2009.09761.pdf) vocoder trained with [LJSpeech](https://keithito.com/LJ-Speech-Dataset/). 
+
+The pre-trained model takes as input a spectrogram and generates a waveform as output. Typically, a vocoder is used after a TTS model that converts an input text into a spectrogram.
+
+The sampling frequency is 22050 Hz.
+
+
+## Install SpeechBrain
+
+```bash
+pip install speechbrain
+```
+
+Please notice that we encourage you to read our tutorials and learn more about
+[SpeechBrain](https://speechbrain.github.io).
+
+### Using the Vocoder as reconstructor
+```python
+import torch
+import torchaudio
+import speechbrain as sb
+from speechbrain.pretrained import DiffWaveVocoder
+from speechbrain.lobes.models.HifiGAN import mel_spectogram
+
+diffwave = DiffWaveVocoder.from_hparams(source="speechbrain/tts-diffwave-ljspeech", savedir="tmpdir")
+
+audio = sb.dataio.dataio.read_audio("speechbrain/tts-diffwave-ljspeech/LJ050-0075.wav")
+audio = torch.FloatTensor(audio)
+audio = audio.unsqueeze(0)
+
+mel = mel_spectogram(
+    sample_rate=22050,
+    hop_length=256,
+    win_length=1024,
+    n_fft=1024,
+    n_mels=80,
+    f_min=0,
+    f_max=8000,
+    power=1.0,
+    normalized=False,
+    norm="slaney",
+    mel_scale="slaney",
+    compression=True,
+    audio=audio,
+)
+
+# Running Vocoder (spectrogram-to-waveform), a fast sampling can be realized by passing user-defined variance schedules. According to the paper, high-quality audios can be generated with only 6 steps (instead of a total of 50).
+waveforms = diffwave.decode_batch(
+    mel,
+    hop_len=256,  # upsample factor, should be the same as "hop_len" during the extraction of mel-spectrogram
+    fast_sampling=True,  # fast sampling is highly recommanded
+    fast_sampling_noise_schedule=[0.0001, 0.001, 0.01, 0.05, 0.2, 0.5],  # customized noise schedule 
+)
+
+torchaudio.save('reconstructed.wav', waveforms.squeeze(1), 22050)
+```
+
+### Using the Vocoder with TTS
+```python
+import torchaudio
+from speechbrain.pretrained import FastSpeech2
+from speechbrain.pretrained import DiffWaveVocoder
+
+# Intialize TTS (FastSpeech2) and Vocoder (DiffWave)
+fastspeech2 = FastSpeech2.from_hparams(source="speechbrain/tts-fastspeech2-ljspeech", savedir="tmpdir_tts")
+diffwave = DiffWaveVocoder.from_hparams(source="speechbrain/tts-diffwave-ljspeech", savedir="tmpdir_vocoder")
+
+input_text = "This is a test run with FastSpeech and DiffWave."
+
+# Running the TTS
+mel_output, durations, pitch, energy = fastspeech2.encode_text(
+  [input_text],
+  pace=1.0,        # scale up/down the speed
+  pitch_rate=1.0,  # scale up/down the pitch
+  energy_rate=1.0, # scale up/down the energy
+)
+
+# Running Vocoder (spectrogram-to-waveform), a fast sampling can be realized by passing user-defined variance schedules. According to the paper, high-quality audios can be generated with only 6 steps (instead of a total of 50).
+waveforms = diffwave.decode_batch(
+    mel_output,
+    hop_len=256,  # upsample factor, should be the same as "hop_len" during the extraction of mel-spectrogram
+    fast_sampling=True,  # fast sampling is highly recommanded
+    fast_sampling_noise_schedule=[0.0001, 0.001, 0.01, 0.05, 0.2, 0.5],  # customized noise schedule 
+)
+
+# Save the waverform
+torchaudio.save('example_TTS.wav',waveforms.squeeze(1), 22050)
+```
+
+### Inference on GPU
+To perform inference on the GPU, add  `run_opts={"device":"cuda"}`  when calling the `from_hparams` method.
+
+### Training
+The model was trained with SpeechBrain.
+To train it from scratch follow these steps:
+1. Clone SpeechBrain:
+```bash
+git clone https://github.com/speechbrain/speechbrain/
+```
+2. Install it:
+```bash
+cd speechbrain
+pip install -r requirements.txt
+pip install -e .
+```
+3. Run Training:
+```bash
+cd recipes/LJSpeech/TTS/vocoder/diffwave/
+python train.py hparams/train.yaml --data_folder /path/to/LJspeech
+```
+You can find our training results (models, logs, etc) [here](https://www.dropbox.com/sh/tbhpn1xirtaix68/AACvYaVDiUGAKURf2o-fvgMoa?dl=0).
+
+
+### Limitations
+The SpeechBrain team does not provide any warranty on the performance achieved by this model when used on other datasets.
+
+# **About SpeechBrain**
+- Website: https://speechbrain.github.io/
+- Code: https://github.com/speechbrain/speechbrain/
+- HuggingFace: https://huggingface.co/speechbrain/
+
+
+# **Citing SpeechBrain**
+Please, cite SpeechBrain if you use it for your research or business.
+
+```bibtex
+@misc{speechbrain,
+  title={{SpeechBrain}: A General-Purpose Speech Toolkit},
+  author={Mirco Ravanelli and Titouan Parcollet and Peter Plantinga and Aku Rouhe and Samuele Cornell and Loren Lugosch and Cem Subakan and Nauman Dawalatabad and Abdelwahab Heba and Jianyuan Zhong and Ju-Chieh Chou and Sung-Lin Yeh and Szu-Wei Fu and Chien-Feng Liao and Elena Rastorgueva and François Grondin and William Aris and Hwidong Na and Yan Gao and Renato De Mori and Yoshua Bengio},
+  year={2021},
+  eprint={2106.04624},
+  archivePrefix={arXiv},
+  primaryClass={eess.AS},
+  note={arXiv:2106.04624}
+}
+```

diffwave.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b434ac56f45486ec899e4ef50fb1bab09443bbe4ccc844acae35486e89541fc2
+size 10582085

hyperparams.yaml

@@ -0,0 +1,44 @@
+# ################################################
+# Basic parameters for a diffwave vocoder
+#
+# Author:
+#  * Yingzhi Wang 2022
+# ################################################
+
+train_timesteps: 50
+beta_start: 0.0001
+beta_end: 0.05
+
+residual_layers: 30
+residual_channels: 64
+dilation_cycle_length: 10
+
+unconditional: False
+
+spec_n_mels: 80
+spec_hop_length: 256
+
+diffwave: !new:speechbrain.lobes.models.DiffWave.DiffWave
+    input_channels: !ref <spec_n_mels>
+    residual_layers: !ref <residual_layers>
+    residual_channels: !ref <residual_channels>
+    dilation_cycle_length: !ref <dilation_cycle_length>
+    total_steps: !ref <train_timesteps>
+    unconditional: !ref <unconditional>
+
+noise: !new:speechbrain.nnet.diffusion.GaussianNoise
+
+diffusion: !new:speechbrain.lobes.models.DiffWave.DiffWaveDiffusion
+    model: !ref <diffwave>
+    beta_start: !ref <beta_start>
+    beta_end: !ref <beta_end>
+    timesteps: !ref <train_timesteps>
+    noise: !ref <noise>
+
+modules:
+    diffwave: !ref <diffwave>
+    diffusion: !ref <diffusion>
+
+pretrainer: !new:speechbrain.utils.parameter_transfer.Pretrainer
+    loadables:
+        diffwave: !ref <diffwave>