Aller au contenu

Palabra TTS v1 Palabra · text-to-speech (TTS) API

Fast enough

109 ms median and 205 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), a tie within 10 ms with Fish Audio Drama 3 Preview and Inworld Realtime TTS-2 on slow turns.

Palabra TTS v1 is a proprietary text-to-speech (TTS) API from Palabra, measured by two independent benchmarks: Coval and Speko.

Speed tier

Fast enough

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

Median time to first audioCoval, 30 days
109 ms
P95 time to first audio (slow turns)Coval, 30 days
205 ms
Naturalness EloSpeko
1520
Listener Elo (95% interval)Artificial Analysis
not rated
Numbers and dates (0 to 1)Speko
0.16
List price per 1M charactersall sources
$12
  • Tagged multilingual by Coval

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

Gradium TTS vs Palabra TTS v1 Fastest TTS APIs, by speed tier

Speed

On Coval's 30-day window ending October 22, 2026 (2,898 runs from US East), Palabra TTS v1 starts speaking after 109.2 ms on a typical turn (median) and 205.3 ms on a slow turn (P95, the slowest one in twenty). That P95 is 9th 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, SpeechifyAI Simba 3.0 and SpeechifyAI Simba 3.2. Speko's board plots it at 86 ms time to first audio from US East; its model page reports a 72 ms sweep.

Voice quality

Speko rates its naturalness at 1,520 Elo from blind A/B votes on phone-agent lines (21st highest of the 43 models Speko rates).

Numbers and dates

Speko's pronunciation test on numbers, dates and currency scores it 0.16 on a 0-to-1 scale (28th highest of the 30 models Speko scores). This is the accuracy measure used here; Coval's word error rate separates models too little to compare them.

Price

Coval lists $12 per 1M characters (since Oct 13, 2026). Speko lists about $12. List prices, as each source prints them.

Benchmark measurements

Latency and accuracy — Coval (30-day window)

Mean TTFA
121.3 ms
P50 / median time-to-first-audio (TTFA)
109.2 ms
P90 TTFA
145.3 ms
P95 TTFA
205.3 ms
Latency consistency (standard deviation)
62.5 ms
Leading silence
4.8 ms
Time to first byte (network arrival)
116.5 ms
Accuracy: word error rate (WER)
3.2%
WER basis
pooled
Latency runs
2,898 runs
Accuracy runs
2,897 runs
Window
30d
Coval snapshot
2026-10-22T14:00:19.256741Z
Coval model ID
palabra-tts-v1
Coval model page
benchmarks.coval.ai
Tagged multilingual
✓
Feature tags
multilingual, voice-cloning
List price per 1M characters
12 USD
Price listed since
Oct 13, 2026

Speko

P50 TTFA, US East
72 ms
Naturalness (Elo)
1,520
Pronunciation robustness (0 to 1)
0.16
Voice drift (lower is steadier)
11
Price per 1M characters
12 USD
Measured on
2026-07-03
Speko model
palabra-tts-v1
Speko model page
benchmarks.speko.ai
Time to first audio, board chart
86 ms

Price (all sources)

Lowest list price per 1M characters
12 USD
Highest list price per 1M characters
12 USD

Not collected

Similar TTS models