Preset speakers
Choose Aiden, Dylan, Eric, Ono_anna, Ryan, Serena, Sohee, Uncle_fu, or Vivian. Serena is the default.
Qwen3 text to speech
Convert text to natural, expressive AI speech in seconds. Perfect for voiceovers, narrations, and custom voices — no recording equipment required.
The same creation, credit, and asset flow as the rest of the studio, with a voice instead of a video.
Choose Aiden, Dylan, Eric, Ono_anna, Ryan, Serena, Sohee, Uncle_fu, or Vivian. Serena is the default.
Leave language on Auto, or set Chinese, English, Japanese, Korean, French, German, Spanish, Portuguese, or Russian.
Attach an optional reference audio clip when you want the reading to follow a specific voice.
Type the line you want spoken. The same text is sent as the prompt and the speech text.
Pick a preset speaker and language, or upload a reference audio clip.
Qwen3 TTS runs through the shared creation history. Finished audio stays with your other generations.
Seedance Art text-to-speech uses Qwen3 TTS. It turns a written line into speech with a preset speaker, a language setting, and an optional reference audio clip.
Voiceovers, narrations, and custom reads use the same credits, uploads, and creation history as Seedance video. Seedance 2.5 and the 2.0 family still generate dialogue inside a video when the shot itself needs synced sound.
Generate up to 30-second clips with larger reference sets and native synced audio.
Generate with Seedance 2.0, 2.0 Fast, and 2.0 Mini in one shared studio workflow.
Switch between Seedance 2.5 and the Seedance 2.0 family in the studio.
Read a practical comparison of duration limits, reference caps, and output quality.
See how the multimodal 2.x family differs from the earlier cinematic baseline.
Review prompt patterns for dialogue, references, and audio-led scenes.
Type the line, choose a Qwen3 speaker, and render the audio.