igna/LuxTTS
1
0
Fork 0
Code Activity

Add audio saving to speech generation in README

Browse Source 
Added audio saving functionality to the speech generation examples.
This commit is contained in:
Yatharth Sharma
2026-01-25 11:08:17 -05:00
committed by GitHub
parent c6efaff7f4
commit a3d45874ac
1 changed files with 10 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
README.md
View File

@@ -45,6 +45,7 @@ lux_tts = LuxTTS('YatharthS/LuxTTS', device='cuda', threads=2) ## change device
#### Simple inference #### Simple inference
```python ```python
import soundfile as sf
from IPython.display import Audio from IPython.display import Audio
text = "Hey, what's up? I'm feeling really great if you ask me honestly!" text = "Hey, what's up? I'm feeling really great if you ask me honestly!"
@@ -58,12 +59,17 @@ encoded_prompt = lux_tts.encode_prompt(prompt_audio, rms=0.01)
## generate speech ## generate speech
final_wav = lux_tts.generate_speech(text, encoded_prompt, num_steps=num_steps) final_wav = lux_tts.generate_speech(text, encoded_prompt, num_steps=num_steps)
## save audio
final_wav = final_wav.numpy().squeeze()
sf.write('output.wav', audio_data, 48000)
## display speech ## display speech
display(Audio(final_wav, rate=48000)) display(Audio(final_wav, rate=48000))
``` ```
#### Inference with sampling params: #### Inference with sampling params:
```python ```python
import soundfile as sf
from IPython.display import Audio from IPython.display import Audio
text = "Hey, what's up? I'm feeling really great if you ask me honestly!" text = "Hey, what's up? I'm feeling really great if you ask me honestly!"
@@ -83,6 +89,10 @@ encoded_prompt = lux_tts.encode_prompt(prompt_audio, rms=rms)
## generate speech ## generate speech
final_wav = lux_tts.generate_speech(text, encoded_prompt, num_steps=num_steps, t_shift=t_shift, speed=speed, return_smooth=return_smooth) final_wav = lux_tts.generate_speech(text, encoded_prompt, num_steps=num_steps, t_shift=t_shift, speed=speed, return_smooth=return_smooth)
## save audio
final_wav = final_wav.numpy().squeeze()
sf.write('output.wav', audio_data, 48000)
## display speech ## display speech
display(Audio(final_wav, rate=48000)) display(Audio(final_wav, rate=48000))
``` ```