R

Silk Mulberry 1.5 text-to-speech model

Rumik

Silk Mulberry 1.5 is a text-to-speech (TTS) model from Rumik. It has no Artificial Analysis score yet. On TTS Arena V2 it scores Elo 1416 (±44, 323 votes).

Key facts

AA Elo–Artificial Analysis
TTS Arena Elo1416±44TTS Arena votes 323
Price (per 1M chars)–USD
Released–Open weights:–
Provider
Rumik
AA rank
–
AA Elo (95% CI)
–
AA samples
–
TTS Arena Elo (uncertainty)
1416 ±44
TTS Arena votes
323
TTS Arena win rate
36.2%
Price (per 1M chars)
–
Released
–
Open weights
–
License
–
Voices tested by AA
–
Languages
–
Voice cloning
–

"–" means the upstream sources do not provide this value; we do not guess.

Nearby models

By TTS Arena Elo; this model highlighted
#ModelAA EloArena EloPrice $/1M charsOpen weights
34Magpie Research Preview NVIDIA–1465––
35NNeuTTS Max Neuphonic–1430––
36TTontaube V1 Tontaube–1426–Open
37RSilk Mulberry 1.5 Rumik–1416––
38FFluxions Vui Fluxions–1405–Open
39MMaya1 Maya Research10481401–Open
40Magpie Multilingual NVIDIA–1397––

Alternatives · Top 5 open-weight by AA Elo

#ModelAA EloArena EloPrice $/1M charsOpen weights
9Fish Audio Drama 3 Preview Fish Audio1248–$15Open
11BBreeze TTS 2 BreezeBlue1230–$34Open
19Ming-3.0-tts Ant Group1191––Open
38Fish Audio S2 Pro Fish Audio1110–$15Open
41Step Audio EditX (Mar 2026) StepFun1100––Open

FAQ

Where does Silk Mulberry 1.5 rank among TTS models?

As of Oct 25, 2026, Silk Mulberry 1.5 has no Artificial Analysis score; its TTS Arena V2 Elo is 1416 (323 votes).

Is Silk Mulberry 1.5 open source?

The sources do not list Silk Mulberry 1.5 as open-weight; it is generally used through the Rumik API.

What is the best Chinese TTS model?

By AA Elo, the top model from a Chinese provider is Qwen-Audio-3.1-TTS-Plus by Alibaba (Elo 1307, #3 overall).

Which top-10 TTS model is cheapest?

Among the AA top 10, the lowest list price is Simba 3.2 (SpeechifyAI) at $6.6 per 1M characters, Elo 1246.

What is the best open-source TTS model?

By AA Elo, the top open-weight model is Fish Audio Drama 3 Preview by Fish Audio (Elo 1248, #9 overall); 18 open-weight models have an AA score.

Sources

"–" means the upstream sources do not provide this value; we do not guess.