Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language: "en"
|
3 |
+
tags:
|
4 |
+
- text-to-speech
|
5 |
+
- TTS
|
6 |
+
- speech-synthesis
|
7 |
+
- Tacotron2
|
8 |
+
- speechbrain
|
9 |
+
license: "apache-2.0"
|
10 |
+
datasets:
|
11 |
+
- LJSpeech
|
12 |
+
metrics:
|
13 |
+
- mos
|
14 |
+
---
|
15 |
+
|
16 |
+
# Sunbird AI Text-to-Speech (TTS) model trained on Luganda text
|
17 |
+
|
18 |
+
### Text-to-Speech (TTS) with Tacotron2 trained on Male Commonvoice Recordings
|
19 |
+
|
20 |
+
This repository provides all the necessary tools for Text-to-Speech (TTS) with SpeechBrain using a [Tacotron2](https://arxiv.org/abs/1712.05884) pretrained on [LJSpeech](https://keithito.com/LJ-Speech-Dataset/).
|
21 |
+
|
22 |
+
The pre-trained model takes in input a short text and produces a spectrogram in output. One can get the final waveform by applying a vocoder (e.g., HiFIGAN) on top of the generated spectrogram.
|
23 |
+
|
24 |
+
|
25 |
+
### Install SpeechBrain
|
26 |
+
|
27 |
+
```
|
28 |
+
pip install speechbrain
|
29 |
+
```
|
30 |
+
|
31 |
+
Please notice that we encourage you to read our tutorials and learn more about
|
32 |
+
[SpeechBrain](https://speechbrain.github.io).
|
33 |
+
|
34 |
+
### Perform Text-to-Speech (TTS)
|
35 |
+
|
36 |
+
```
|
37 |
+
import torchaudio
|
38 |
+
from speechbrain.pretrained import Tacotron2
|
39 |
+
from speechbrain.pretrained import HIFIGAN
|
40 |
+
|
41 |
+
# Intialize TTS (tacotron2) and Vocoder (HiFIGAN)
|
42 |
+
tacotron2 = Tacotron2.from_hparams(source="speechbrain/tts-tacotron2-ljspeech", savedir="tmpdir_tts")
|
43 |
+
hifi_gan = HIFIGAN.from_hparams(source="speechbrain/tts-hifigan-ljspeech", savedir="tmpdir_vocoder")
|
44 |
+
|
45 |
+
# Running the TTS
|
46 |
+
mel_output, mel_length, alignment = tacotron2.encode_text("Mary had a little lamb")
|
47 |
+
|
48 |
+
# Running Vocoder (spectrogram-to-waveform)
|
49 |
+
waveforms = hifi_gan.decode_batch(mel_output)
|
50 |
+
|
51 |
+
# Save the waverform
|
52 |
+
torchaudio.save('example_TTS.wav',waveforms.squeeze(1), 22050)
|
53 |
+
```
|
54 |
+
|
55 |
+
If you want to generate multiple sentences in one-shot, you can do in this way:
|
56 |
+
|
57 |
+
```
|
58 |
+
from speechbrain.pretrained import Tacotron2
|
59 |
+
tacotron2 = Tacotron2.from_hparams(source="speechbrain/TTS_Tacotron2", savedir="tmpdir")
|
60 |
+
items = [
|
61 |
+
"A quick brown fox jumped over the lazy dog",
|
62 |
+
"How much wood would a woodchuck chuck?",
|
63 |
+
"Never odd or even"
|
64 |
+
]
|
65 |
+
mel_outputs, mel_lengths, alignments = tacotron2.encode_batch(items)
|
66 |
+
|
67 |
+
```
|
68 |
+
|
69 |
+
### Inference on GPU
|
70 |
+
To perform inference on the GPU, add `run_opts={"device":"cuda"}` when calling the `from_hparams` method.
|