Rime Alternatives for Voice Agent TTS Compared (2026)
Rime prices its two voice families by usage: Mist v3 at $0.03 per 1K characters (about $0.03 a minute, Rime says), built for the lowest latency, and Coda at $0.05 per 1K for the most natural expression. Starter needs no credit card and allows 20 concurrent generations; Enterprise adds volume pricing, on-prem or VPC deployment, unlimited custom voice clones and a BAA. The alternatives below: ElevenLabs at $0.04 per 1K characters for Flash / v4 Turbo ($0.08 for v4, v3 and v2 Multilingual), Cartesia's monthly credit plans (Free with 20K credits, Pro at $5 for 100K, Startup at $49 for 1.25M, Scale at $299 for 8M; one credit is one character), Deepgram Aura-2 at $0.030 per 1K (Flux TTS $0.045), Speechmatics at $0.011 per 1K with the first 1M characters free, and Soniox at about $0.70 per hour of generated speech. Prices are from each vendor's own pricing page.
Rime prices per 1K characters and says that's about a minute of audio. Put every option in the same unit before you compare: per 1K characters, or per hour of speech.
ElevenLabs, Cartesia, Deepgram, Speechmatics and Soniox compared with Rime on published text-to-speech pricing, free allowances and languages.
ElevenLabs Agents- Starting price
- Agents: $0.08 per minute
- Best for
- Builders who want ElevenLabs' voices and speech models in one agent pipeline
Cartesia- Starting price
- Free; paid from $5/mo
- Best for
- Text-to-speech from a vendor focused on low latency (its own figures)
Deepgram- Starting price
- Nova-3 streaming $0.0048/min ($0.29/hr); $200 free credit
- Best for
- Streaming speech-to-text for voice agents
Speechmatics- Starting price
- From $0.24/hr pay as you go (as low as ~$0.12/hr with discounts)
- Best for
- Multilingual transcription with language switching
Soniox- Starting price
- ~$0.10/hr async; ~$0.12/hr real-time
- Best for
- Very low-cost speech-to-text
TTS models compared
- Price
- $0.04 / 1K chars
- Latency
- ~75 ms (model inference)
- Languages
- 32
- Cloning
- Yes
- Price
- $0.04 / 1K chars
- Latency
- ~100 ms median (model inference)
- Languages
- 90+
- Cloning
- Yes
- Price
- Free tier; from $5/mo (100K credits)
- Latency
- —
- Languages
- 44
- Cloning
- Yes (instant, paid plans)
- Price
- $0.03 / 1K chars (~$0.03/min)
- Latency
- 37 ms P50 time-to-first-audio
- Languages
- 4 (en, fr, de, es)
- Cloning
- Enterprise
- Price
- $0.05 / 1K chars (~$0.05/min)
- Latency
- 96 ms P50 time-to-first-audio
- Languages
- 8
- Cloning
- Enterprise
- Price
- $0.030 / 1K chars
- Latency
- —
- Languages
- 7 (en, es, de, fr, nl, it, ja)
- Cloning
- —
- Price
- $0.045 / 1K chars
- Latency
- —
- Languages
- —
- Cloning
- —
- Price
- $0.011 / 1K chars; first 1M chars free
- Latency
- —
- Languages
- —
- Cloning
- —
- Price
- ~$0.70 / hr of speech
- Latency
- Real-time streaming
- Languages
- —
- Cloning
- —
See all voice models · Latency and prices as each vendor publishes them.
What to check before you choose
- Compare in one unit. Using Rime's own rule of thumb (1K characters is about a minute), Mist v3 works out to about $1.80 an hour of speech and Coda about $3.00, against Soniox at about $0.70 an hour. Cartesia counts 750–800 characters (credits) per minute, a little under Rime's 1K, so per-minute figures from different vendors aren't exactly alike.
- Check the free allowance twice. Rime's Starter card says about 800 free minutes (about 800K characters), but the FAQ on the same page says every new account starts with 3,000 free minutes. Plan on 800 until Rime confirms.
- Need word timings for interruptions? Rime lists word-level timestamps for Coda but not for Mist v3, so check this before you pick the faster model.
- Languages differ a lot. Rime lists 4 production languages for Mist v3 (English, French, German, Spanish) and 8 for Coda; Deepgram Aura-2 covers 7; ElevenLabs v4 Turbo lists 90+.
- Treat latency figures as vendor claims. Rime lists 37 ms median time to first audio for Mist v3 and 96 ms for Coda; ElevenLabs says about 75 ms for Flash and about 100 ms for v4 Turbo. Measure from your own servers before you decide.
- Voice cloning and BAAs: Rime's custom voice clones and BAA are on Enterprise; Cartesia includes instant cloning from Pro ($5/mo) and professional cloning from Startup ($49/mo), with BAAs on Enterprise; ElevenLabs lists BAAs on Enterprise.
- Test on free credit. ElevenLabs' free plan includes 20,000 Flash / v4 Turbo characters a month, Speechmatics gives the first 1M characters free, Cartesia's Free plan has 20K credits a month, and Deepgram's $200 credit also covers text-to-speech. Through Dec 31, 2026, Deepgram also matches Flux TTS spend with credits, up to $500.
Not sure which fits? Answer a few quick questions in the finder →
Prices from vendors' official pricing pages, checked October 20, 2026. Not sponsored; order isn't a ranking.