Chinese audio for less than 15 seconds

[HaSaKiYasuooo]

I used a Chinese speech model to generate audio, and the generated audio can never exceed 15 seconds. How can I solve this problem?

[anacondabitch]

can you tell me how was the quality of the audio generated by model

[C0untFloyd]

With chinese model you probably mean a chinese speaker and for generation you used Swap Voice? Otherwise there is no 15s limitation in this fork?!?

[HaSaKiYasuooo]

你能告诉我模型生成的音频质量如何吗

Bark good at English but terrible at other languages

[HaSaKiYasuooo]

With chinese model you probably mean a chinese speaker and for generation you used Swap Voice? Otherwise there is no 15s limitation in this fork?!?

So how do I solve the loss of text in the audio when Chinese is generating speech

[C0untFloyd]

Are you sure you're using this fork? As I wrote previously I don't have a audio limit.

Bark good at English but terrible at other languages

I disagree, you just have to find a good speaker voice. At least for the languages I know, it might be different for chinese 😏

[HaSaKiYasuooo]

I tried many times to use Chinese for audio output, but the results are not satisfactory, Could you please give me some solutions

[C0untFloyd]

Is it this problem? https://github.com/suno-ai/bark/issues/324