Soniox · Proprietary · Read 2026-10-27
Soniox TTS Real-Time v2: latency, voice quality, price and languages
- Rank
- 10 (tied) / 23
- Speed tier
- Visible delay
- P95
- 324.7 ms
- Price / 1M chars
- $13 to $14.81
Soniox TTS Real-Time v2 is a proprietary text-to-speech model from Soniox. Coval measures a median of 264.4 ms and a P95 of 324.7 ms to first audio, which puts it in the Visible delay speed tier, rank 10 (tied) of 23 in the multilingual ranking. Artificial Analysis or Speko tested it in 10 languages.
This summary is generated from the measures below; it is not a vendor description. Measures a source does not hold are shown as n/a and are never estimated.
Speed (Coval, 30 days, US East, English)
| Speed tier | Visible delay |
| Median time to first audio | 264.4 ms |
| P90 | 311.3 ms |
| P95 | 324.7 ms |
| Mean | 260.3 ms |
| Standard deviation | 47.9 ms |
| Calls in the window | 2,365 |
| Cekura median, streamed text | 183 ms |
| Cekura median, full text | 241 ms |
Voice quality (never added together)
| Listener Elo (Artificial Analysis) | 1,180 ±11 |
| Customer-service Elo (Artificial Analysis) | 1,213 ±21 |
| Naturalness (Speko) | 1,606 |
| Naturalness estimate (Cekura, 1 to 5) | 4.37 |
Accuracy
| Reading accuracy (Cekura) | 83.6% |
| Numbers (Cekura) | 78% |
| Codes (Cekura) | 86% |
| Context (Cekura) | 88% |
| Medical terms (Cekura) | 97% |
| Hard passages judged correct (Artificial Analysis) | 78.8% |
| Numbers and dates (Speko score) | 0.49 |
| Word error rate (Coval) | 2.6% |
Price per 1 million characters
| Range across sources | $13 to $14.81 |
| Artificial Analysis | $14.20 |
| Speko | $13 |
| Coval | $14.10 |
| Cekura | $14.81 |
Coverage
| Access | Proprietary |
| Streaming | Measured |
| Coval tags | multilingual, voice-cloning |
| Rated for | Knowledge sharing, Assistants, Entertainment, Customer service |
| Languages tested | Arabic, Chinese, Filipino, French, German, Hindi, Japanese, Norwegian, Spanish, Thai |
Not collected
Every measure of the database is filled for this model.
Sources
- https://artificialanalysis.ai/text-to-speech (read 2026-10-27)
- https://artificialanalysis.ai/text-to-speech/leaderboard/provider-voice (read 2026-10-27)
- https://benchmarks.cekura.ai/text-to-speech (read 2026-10-27)
- https://benchmarks-api-6wxgp27p2a-ue.a.run.app/v1/results/aggregates?benchmark=TTS&window=30d&include_series=false (read 2026-10-27)
- https://benchmarks.speko.ai/tts/ (read 2026-10-27)
- Vendor site: https://soniox.com/
Keep exploring
Frequently asked questions
- How fast is Soniox TTS Real-Time v2?
- Coval measures a median time to first audio of 264.4 ms and a P95 of 324.7 ms for Soniox TTS Real-Time v2, over 30 days from the US East in English (Visible delay tier).
- How much does Soniox TTS Real-Time v2 cost?
- The benchmarks print a list price of $13 to $14.81 per million characters for Soniox TTS Real-Time v2.
- Which languages was Soniox TTS Real-Time v2 tested in?
- Artificial Analysis or Speko tested Soniox TTS Real-Time v2 in Arabic, Chinese, Filipino, French, German, Hindi, Japanese, Norwegian, Spanish, Thai.