Audio Conversion Extension Module
For the general public, please refer to README_normal.md.
Apply basic voice pack1.
imouto = kawaii_voice('voice.mp3')
result = imouto.voice_pack2()
Apply pitch change.
imouto = kawaii_voice('voice.mp3')
result = imouto.pitch(2.0)
Note: Audio data is passed as pydub.AudioSegment.
AudioSegment. You can also convert it with numpy as follows
np_array = numpy.array(imouto.audio.get_array_of_samples())
| function | outline |
|---|---|
| formant (value, f0_rate) | The higher the value, the closer it is to a woman's voice. |
| speed (value) | Change play speed. |
| pitch (value) | Change audio pitch. |
| volume (value) | Change audio volume. |
| bass_boost (option) | Bass boost using a low-pass filter. |
| normalize (option) | Normalize audio data. |
| voice_pack1 () | Audio conversion (old ver.) |
| voice_pack2 () | Audio conversion tuned for gTTS(ja) |
| music_pack1 () | Nightcore conversion. |
| Voice source | option | source | result |
|---|---|---|---|
| gTTS | voice_pack2 | voice_gtts.mp3 | voice_gtts_pack2.mp3 |
| pyttsx3 | formant(1.2, 1.7) | voice_pyttsx3.mp3 | voice_pyttsx3_formant.mp3 |
| Sample code | |
|---|---|
| example_voice.py | gTTS ver. |
| example_voice2.py | pyttsx3 ver. |
| example_music.py | convert song to Nightcore. |
It's still just a basic voice converter, but I'm hoping to add advanced voice quality conversion as well.
README紳士向け版を最後まで読んでくれるなんてやさしいね！ ありがとう お兄ちゃん！