Qwen3-TTS 0.6B fast — API
generate_custom_voice
+
stream_custom_voice
Text
Language
French
Speaker
Serena
Instruct
Model size
0.6B
Audio
Chunks WAV base64 (stream)
Générer
Stream