Sprag
Sprag provides speech-to-speech agents and an STT-to-LLM-to-TTS voice pipeline through one API.
What Sprag is
Sprag provides speech-to-speech agents and an STT-to-LLM-to-TTS voice pipeline through one API. Rhythm transcribes audio in 30 languages for streaming or batch use. Chorus synthesizes speech with built-in, cloned or designed voices, while Symphony supports real-time tool calling, interruptions and turn-taking. Symphony Notes describes audio clips, including speech, speakers, emotion and background sound. Sprag serves its speech models through an OpenAI-shaped HTTP API, so existing clients can connect by changing the base URL and API key. It documents an integration with Pipecat and lists LiveKit in its integrations navigation. The service is available through web, API and a self-hosted deployment option; listed export formats include MP3, Opus, AAC, FLAC, WAV and PCM. Sprag has no free plan. Chorus is listed at $0.70 per month, billed at $0.70 per hour of speech generated, and other options cover voice design, cloning, transcription, realtime agents and captioning. Realtime sessions for Rhythm and Symphony cost 25% above batch rates.
Good
- One API covers speech agents and a speech pipeline.
- Rhythm supports streaming and batch transcription.
- Chorus offers built-in, cloned and designed voices.
- API uses an OpenAI-shaped HTTP interface.
- Multiple audio export formats are listed.
Watch out
- There is no free plan.
- Realtime Rhythm and Symphony sessions cost 25% above batch rates.
- Chorus Clone adds a charge for reference audio.
Who it suits
Sprag is intended for developers aged 16 and older who need speech recognition, synthesis or realtime voice agents. Its API and self-hosting options may suit teams integrating speech capabilities into their own applications or deployments.
Verdict
Sprag combines transcription, voice generation and realtime agent capabilities behind an API, with web and self-hosted options also listed. Review the distinct usage rates, especially the realtime premium and voice-cloning charges, before choosing a plan.
Plans and prices
From Sprag's own pricing page, checked 20 Sep 2026.
| Plan | Price | What it gives you |
|---|---|---|
| Chorus | $0.70 a month | Stock-voice text-to-speech · $0.70 per hour of speech generated; also listed as $19.44 per 1M audio tokens |
| Chorus Design | $15 a month | Designs new voices from a text prompt |
| Chorus Clone | $4.50 a month | Clones a voice from a short reference · Plus $1.00 per hour of input audio |
| Symphony Notes | Free | Audio captioning · + $2.03 per 1M output tokens · $0.0019/ minute |
| Rhythm | $0.08, paid once | Streaming and batch · realtime sessions cost 25% over batch rates · $0.075/ hour of audio submitted; batch rate |
| Symphony | $0.09, paid once | Text, image, audio, and video in · text and speech out · tool calling · realtime sessions cost 25% over batch rates · $0.09/ hour of audio; batch rate |
| Chorus — built-in voice | $0.70, paid once | Built-in voices · $19.44 per 1M audio tokens · $0.70/ hour of speech generated |
| Chorus — cloned voice | $4.50, paid once | $0.27 per hour of speech · + $1.00 per hour of reference audio · $4.50/ 1M characters |
| Chorus — designed voice | $15, paid once | Voice design · $15.00/ 1M characters |
The facts
Each one read from Sprag's own pages, with the date it was checked.
What it does
- API access
- Yeschecked 20 Sep 2026
- API compatibility
- Sprag serves speech models through an OpenAI-shaped HTTP API, and existing clients can point to it by changing the base URL and API key.checked 7 Oct 2026
- Audience
- Sprag says its products and services are intended for developers aged 16 and older.checked 7 Oct 2026
- Captioning
- Symphony Notes describes audio clips, including speech, speakers, emotion, and background sound.checked 7 Oct 2026
- Enterprise
- The pricing page says users can contact Sprag for an organization account for enterprise use.checked 7 Oct 2026
- Export formats
- mp3,opus,aac,flac,wav,pcmchecked 20 Sep 2026
- Integrations
- Sprag documents integration with Pipecat using its OpenAI services and lists LiveKit in its integrations navigation.checked 7 Oct 2026
- Languages
- 10checked 20 Sep 2026
- Limit
- The pricing page says realtime sessions bill at 25% above batch rates for Rhythm and Symphony, while speech output costs the same either way.checked 7 Oct 2026
- Privacy and security
- Sprag says data is stored in its cloud infrastructure, transfers use HTTPS/TLS, and payment card details are processed by Stripe rather than stored on its systems.checked 7 Oct 2026
- Privacy controls
- Sprag says it does not sell personal information and provides California residents an opt-out for advertising-measurement cookies that may constitute sharing.checked 7 Oct 2026
- Product
- Sprag offers native speech-to-speech agents and an STT-to-LLM-to-TTS voice pipeline through one API.checked 7 Oct 2026
- Realtime agents
- Symphony supports realtime tool calling and natural interruptions and turn-taking.checked 7 Oct 2026
- Self-hosting
- The homepage links to a self-hosted deployment option.checked 7 Oct 2026
- Speech recognition
- Rhythm transcribes audio in 30 languages and supports streaming and batch use.checked 7 Oct 2026
- Speech synthesis
- Chorus synthesizes speech using a built-in, cloned, or designed voice.checked 7 Oct 2026
- Support
- The pricing page directs users with questions about account costs, billing, rate limits, enterprise agreements, and metering to contact Sprag.checked 7 Oct 2026
- Voice cloning
- Yeschecked 20 Sep 2026
Money
- Free plan
- Nochecked 20 Sep 2026
Where it runs
- Platforms
- web,apichecked 20 Sep 2026
Alternatives to Sprag
- 1
Voicebox
Free planOpen sourceAndroidiOSWebAPI9.3From $12 a year - 2
Speechify
Free planAndroidiOSWebAPI9.0From $10 a month - 3
Resemble AI
Free planAndroidiOSWebAPI9.0From $280 a month - 4
iSpeech
Free planOpen sourceAndroidiOSWebAPI9.0Free plan - 5
Fliki
Free planAndroidiOSWebAPI8.8From $28 a month - 6
Voice.ai
Free planAndroidiOSWebAPI8.8From $5 a month