search

suno-ai / bark

🔊 Text-Prompted Generative Audio Model
MIT License
36.23k stars 4.26k forks source link

Voice changes too much #577

Open geb6426 opened 5 months ago

geb6426 commented 5 months ago

They aren't same person's voice everytime. How to change it to one person's voice instead of a different voice everytime python -m bark --text "Hello, my name is Suno." --output_filename "example.wav"

JonathanFly commented 5 months ago

You need to specify a voice, or a history prompt. 

python -m bark --text "Hello, my name is Suno and I am voice 6." --history_prompt 'v2/en_speaker_6' --output_filename "06_example.wav"
PhilipAmadasun commented 1 month ago

@JonathanFly How do I do this in python code. Doe the "geenrate_audio" function have this as an argument?

vieirinhasantana commented 2 weeks ago

@PhilipAmadasun There is a parameter you can pass to set the voice.

audio_array = generate_audio("TEXT_PROMPT", history_prompt="v2/en_speaker_6" )

The voices available here. https://suno-ai.notion.site/8b8e8749ed514b0cbf3f699013548683?v=bc67cff786b04b50b3ceb756fd05f68c

JonathanFly commented 2 weeks ago

@PhilipAmadasun There is a parameter you can pass to set the voice.

audio_array = generate_audio("TEXT_PROMPT", history_prompt="v2/en_speaker_6" )

The voices available here. https://suno-ai.notion.site/8b8e8749ed514b0cbf3f699013548683?v=bc67cff786b04b50b3ceb756fd05f68c

This is correct, but there are many other voices you can use. You can save any random Bark output as a new voice .npz file and use it in the future.

PhilipAmadasun commented 1 week ago

@JonathanFly Are there parameters to make the model run faster for real time?