Models
Voice models compared
Vendor-stated figures only, checked October 14–15, 2026. A dash means the vendor doesn't publish it.
Find my voice model →Reach teams choosing voice AI
Sponsor this spot. Email Thomas@orilo.io
Text-to-speech
ElevenLabs
- Latency
- ~75 ms (model inference)
- Price
- $0.04 / 1K chars
- Languages
- 32
- Cloning
- Yes
ElevenLabs
- Latency
- ~100 ms median (model inference)
- Price
- $0.04 / 1K chars
- Languages
- 90+
- Cloning
- Yes
Cartesia
- Latency
- —
- Price
- Free tier; from $5/mo (100K credits)
- Languages
- 44
- Cloning
- Yes (instant, paid plans)
Rime
- Latency
- 37 ms P50 time-to-first-audio
- Price
- $0.03 / 1K chars (~$0.03/min)
- Languages
- 4 (en, fr, de, es)
- Cloning
- Enterprise
Rime
- Latency
- 96 ms P50 time-to-first-audio
- Price
- $0.05 / 1K chars (~$0.05/min)
- Languages
- 8
- Cloning
- Enterprise
Deepgram
- Latency
- —
- Price
- $0.030 / 1K chars
- Languages
- 7 (en, es, de, fr, nl, it, ja)
- Cloning
- —
Speechmatics
- Latency
- —
- Price
- $0.011 / 1K chars; first 1M chars free
- Languages
- —
- Cloning
- —
Speech-to-text
Deepgram
- Latency
- Built-in turn detection
- Price
- $0.0065/min (English)
- Languages
- English or multilingual
- Cloning
- —
ElevenLabs
- Latency
- ~150 ms partials
- Price
- $0.39 / hr
- Languages
- 90+
- Cloning
- —
Deepgram
- Latency
- —
- Price
- $0.0048/min streaming (monolingual, promo; reg. $0.0077)
- Languages
- 45+ (Nova models)
- Cloning
- —
AssemblyAI
- Latency
- Async (pre-recorded)
- Price
- $0.21 / hr
- Languages
- 18
- Cloning
- —
- Latency
- Live transcription
- Price
- $0.017/min
- Languages
- —
- Cloning
- —
Speechmatics
- Latency
- Built-in turn detection and VAD (real-time)
- Price
- $0.30 / hr real-time
- Languages
- 55+
- Cloning
- —
Speechmatics
- Latency
- Real-time or files
- Price
- $0.40 / hr real-time; $0.24 / hr files
- Languages
- 56, switches mid-sentence
- Cloning
- —
Soniox
- Latency
- Real-time streaming
- Price
- ~$0.12 / hr real-time; ~$0.10 / hr async
- Languages
- 60+ (translation included)
- Cloning
- —
Gladia
- Latency
- Async (files)
- Price
- €0.16 / hr async (real-time €0.68 / hr)
- Languages
- 100+
- Cloning
- —
Speech-to-speech
OpenAI
- Latency
- —
- Price
- Audio $32 in / $64 out per 1M tokens
- Languages
- —
- Cloning
- —
- Latency
- —
- Price
- Audio $10 in / $20 out per 1M tokens
- Languages
- —
- Cloning
- —
OpenAI
- Latency
- —
- Price
- $0.05/min, billed per second (backend model extra)
- Languages
- —
- Cloning
- —
- Latency
- —
- Price
- Free tier; audio ~$0.005/min in, ~$0.018/min out
- Languages
- —
- Cloning
- —
ElevenLabs latency excludes application and network time. Samples open on each vendor’s own demo page; we don’t rehost audio. PlayHT omitted: no verifiable current pricing.
Calculator
Estimate your monthly cost
Uses only models whose verified price is quoted per minute or per hour. List prices only: no discounts, taxes, phone or LLM costs.
| Model | Type | Verified price | Est. / month |
|---|---|---|---|
| Soniox stt-rt-v5Soniox · STT · ~$0.12 / hr real-time; ~$0.10 / hr async | STT | ~$0.12 / hr real-time; ~$0.10 / hr async | ≈$20.00 |
| Gladia Solaria-3Gladia · STT · €0.16 / hr async (real-time €0.68 / hr) | STT | €0.16 / hr async (real-time €0.68 / hr) | €26.67 |
| AssemblyAI Universal-3.5 ProAssemblyAI · STT · $0.21 / hr | STT | $0.21 / hr | $35.00 |
| Deepgram Nova-3Deepgram · STT · $0.0048/min streaming (monolingual, promo; reg. $0.0077) | STT | $0.0048/min streaming (monolingual, promo; reg. $0.0077) | $48.00 |
| Speechmatics Linden 1Speechmatics · STT · $0.30 / hr real-time | STT | $0.30 / hr real-time | $50.00 |
| Deepgram FluxDeepgram · STT · $0.0065/min (English) | STT | $0.0065/min (English) | $65.00 |
| ElevenLabs Scribe v2 RealtimeElevenLabs · STT · $0.39 / hr | STT | $0.39 / hr | $65.00 |
| Speechmatics Melia 1Speechmatics · STT · $0.40 / hr real-time; $0.24 / hr files | STT | $0.40 / hr real-time; $0.24 / hr files | $66.67 |
| Soniox TTSSoniox · TTS · ~$0.70 / hr of speech | TTS | ~$0.70 / hr of speech | ≈$117 |
| OpenAI gpt-realtime-whisperOpenAI · STT · $0.017/min | STT | $0.017/min | $170 |
| OpenAI GPT-Live 1OpenAI · Speech-to-speech · $0.05/min, billed per second (backend model extra) | Speech-to-speech | $0.05/min, billed per second (backend model extra) | $500 |
Hourly prices are divided by 60. “≈” means the vendor quotes an approximate rate. Euro prices are shown in euros, not converted.