Aller au contenu

SpeechifyAI Simba 3.0 Speechify · text-to-speech (TTS) API

Fast enough

101 ms median and 148 ms P95 time to first audio on Coval (30 days to October 22, 2026): Fast enough speed tier.

Benchmark summary: Fast enough speed tier on Coval's P95 (fast enough for most agents).

SpeechifyAI Simba 3.0 is a proprietary text-to-speech (TTS) API from Speechify, measured by two independent benchmarks: Artificial Analysis and Coval.

Speed tier

Fast enough

P95 from 138 ms to 263 ms. Fast enough for most agents.

Median time to first audioCoval, 30 days
101 ms
P95 time to first audio (slow turns)Coval, 30 days
148 ms
Naturalness EloSpeko
not rated
Listener Elo (95% interval)Artificial Analysis
1,121 ±11
Numbers and dates (0 to 1)Speko
not rated
List price per 1M charactersall sources
$6.6 to $10
  • Tagged multilingual by Coval
  • Customer service (Artificial Analysis): 43rd, Elo 1132 ±26

Data read October 22, 2026. No combined score: methodology

Gradium TTS vs SpeechifyAI Simba 3.0 Fastest TTS APIs, by speed tier

Speed

On Coval's 30-day window ending October 22, 2026 (3,371 runs from US East), SpeechifyAI Simba 3.0 starts speaking after 101.4 ms on a typical turn (median) and 148.1 ms on a slow turn (P95, the slowest one in twenty). That P95 is 6th lowest of the 32 models Coval measures, which puts it in the Fast enough tier: fast enough for most agents. On the median it is a tie (gap under 10 ms) with Alibaba Qwen3 TTS 1.7b, Palabra TTS v1 and SpeechifyAI Simba 3.2.

Voice quality

In the Artificial Analysis Speech Arena it holds 1,121 ±11 Elo (95% interval), 35th highest of the 98 models Artificial Analysis rates; scores whose intervals overlap are a tie. In its Customer Service category, Artificial Analysis places it 43rd, at 1,132 Elo.

Numbers and dates

Speko has not scored its pronunciation of numbers and dates.

Price

Coval lists $10 per 1M characters (since Sep 22, 2026). Artificial Analysis lists $6.6. List prices, as each source prints them: a range of $6.6 to $10.

Languages tested

Tested in Spanish, German, French and Portuguese by Artificial Analysis or Speko. A language missing here was not tested, which does not mean it is unsupported.

Benchmark measurements

Latency and accuracy — Coval (30-day window)

Mean TTFA
110.4 ms
P50 / median time-to-first-audio (TTFA)
101.4 ms
P90 TTFA
133.7 ms
P95 TTFA
148.1 ms
Latency consistency (standard deviation)
67.5 ms
Leading silence
11.3 ms
Time to first byte (network arrival)
99.1 ms
Accuracy: word error rate (WER)
1.9%
WER basis
pooled
Latency runs
3,371 runs
Accuracy runs
3,372 runs
Window
30d
Coval snapshot
2026-10-22T14:00:19.256741Z
Coval model ID
simba-3.0
Coval model page
benchmarks.coval.ai
Tagged multilingual
✓
Feature tags
emotion-control, multilingual, voice-cloning
List price per 1M characters
10 USD
Price listed since
Sep 22, 2026

Artificial Analysis Speech Arena

Customer service rank
43
Customer service Elo
1,132
Customer service Elo 95% interval (±)
26
Speech Arena Elo (blind listener preference)
1,121
Elo 95% interval (±)
11
Arena rank
35
Arena rank range
33-38
Arena appearances
2,997
Voices tested
8
Price per 1M characters
6.6 USD
Released
Feb 2026
Artificial Analysis model
Simba 3.0
Artificial Analysis leaderboard
artificialanalysis.ai

Price (all sources)

Lowest list price per 1M characters
6.6 USD
Highest list price per 1M characters
10 USD

Not collected

Similar TTS models