Freedom251 › Text-to-speech tools
Speechbase
by Speechbase
What Speechbase is
Speechbase is an audio platform for AI teams, available through the web and an API. It combines text-to-speech provider access with voice management, pronunciation controls, moderation rules, and request monitoring. Its gateway offers one API for 17 providers, including OpenAI, ElevenLabs, Cartesia, Google, Deepgram, and Hume. Teams can save voices for reuse across providers and models, create dialogue from different voices or providers, and receive normalized audio with per-turn timestamps. Speech synthesis returns word-level timestamps that can be converted to SRT or WebVTT captions. Logs record provider, model, voice, status, latency, character counts, options, errors, and moderation decisions. Exports are MP3, WAV, and PCM. Speechbase lists a free plan with 5,000 credits and Pro at $25 a month, billed annually at $300. Its open-source SDK is Apache 2.0 licensed.
Who it suits
It suits AI teams building speech features across several text-to-speech providers. Teams needing shared voices, captions, pronunciation controls, or request logs may find its tools relevant.
What it does well
- One API across 17 speech providers
- Word-level timestamps support caption conversion
- Dialogue can combine voices and providers
- Central pronunciation and moderation controls
- MP3, WAV, and PCM exports
What to know first
- Free plan has 5,000 credits
- No free trial is listed
- Privacy policy allows third-party AI sharing
- Published voice limits conflict between tables
Plans
| Plan | Billed and limits | Price |
|---|---|---|
| Free | Free forever — no credit card · 5,000 credits · 1 day request log retention · 5 pronunciation entries · 5 saved voices · 1 team seat | Free |
| Pro | Billed annually at $300 · 30,000 credits · 30 days request log retention · moderation rules · unlimited pronunciation entries · 100 saved voices · 5 team seats | $25 a month |
| Enterprise | Annual contract · custom rates · Unlimited credits, voices, and seats · SAML / SSO + SCIM · bring-your-own object storage · invoice / PO billing · dedicated Slack channel | On request |
As speechbase.ai lists them, read 11 Oct 2026. The maker's own page has the final word.
Verdict
Speechbase brings provider routing, voice workflows, captions, and observability into one service. Check the plan details and data-sharing terms before choosing it for production speech work.
Facts from Speechbase's own pages
- AI data processing
- The privacy policy says input, output, and personal information may be shared with third-party AI providers to provide AI features. · speechbase.ai, 11 Oct 2026
- API access
- Yes · speechbase.ai, 20 Sept 2026
- Company and headquarters
- Speechbase is operated by Jellypod, Inc., which lists its headquarters as 2261 Market Street, STE 22258, San Francisco, California 94114. · speechbase.ai, 11 Oct 2026
- Dialogue generation
- Speechbase can combine turns from different voices, models, or providers into one audio file with normalized volume and per-turn timestamps. · speechbase.ai, 11 Oct 2026
- Export formats
- mp3,wav,pcm · speechbase.ai, 20 Sept 2026
- Free plan
- Yes · speechbase.ai, 20 Sept 2026
- Headquarters
- San Francisco, California, United States · speechbase.ai, 28 Sept 2026
- Observability
- Request logs include provider, model, voice, status, latency, character counts, provider options, errors, and moderation decisions. · speechbase.ai, 11 Oct 2026
- Platforms
- web,api · speechbase.ai, 20 Sept 2026
- Pricing discrepancy
- The Free card lists 5 saved voices and 1 seat while the comparison table labels the same tier Hobby with 10 saved voices; the Pro card lists 100 saved voices and 5 seats while the comparison table says unlimited for both. · speechbase.ai, 11 Oct 2026
- Pricing limits
- The pricing comparison lists standard rate limits for Hobby, higher limits for Pro, and custom limits for Enterprise. · speechbase.ai, 11 Oct 2026
- Pronunciations and moderation
- Centralized pronunciation dictionaries and workspace moderation rules can apply across providers and models. · speechbase.ai, 11 Oct 2026
- Provider integrations
- Its gateway provides one API across 17 text-to-speech providers, including OpenAI, ElevenLabs, Cartesia, Google, Deepgram, and Hume. · speechbase.ai, 11 Oct 2026
- Purpose
- Speechbase describes itself as an audio platform for AI teams that combines a TTS gateway, observability, pronunciation tools, and voice management. · speechbase.ai, 11 Oct 2026
- SDK
- The site describes its open-source SDK as Apache 2.0 licensed and says it runs anywhere. · speechbase.ai, 11 Oct 2026
- Security and privacy
- The privacy policy says the company uses technical and organizational security measures but does not guarantee that data transmission or storage is 100% secure. · speechbase.ai, 11 Oct 2026
- Paid from
- 25 · speechbase.ai, 20 Sept 2026
- Support
- The privacy policy says users can contact the company for inquiries and support, and the site provides sales contact at [email protected]. · speechbase.ai, 11 Oct 2026
- Timestamps
- The site says it returns word-level timestamps for every synthesis and supports conversion to SRT or WebVTT captions. · speechbase.ai, 11 Oct 2026
- Voice cloning
- Yes · speechbase.ai, 20 Sept 2026
- Voice library
- Users can save voices and reuse them across providers and models by alias or ID; voice cloning is marked as coming soon. · speechbase.ai, 11 Oct 2026
Where Speechbase stands on Freedom251
- Text-to-speech tools#91 of 232
- Text-to-Speech Software#121 of 223