# Speech APIs compared 2026: text-to-speech and transcription rates

URL: https://infro.io/blog/speech-apis-compared-2026
Author: INFRO Editorial (https://infro.io/blog/editorial)
Publisher: INFRO
Published: 2026-10-20
Last modified: 2026-10-20
Last reviewed: 2026-10-20 (next review due 2026-11-19)
Topic: Agents and multimodal systems
Format: Comparison

> In October 2026, text-to-speech costs $15 per million characters on OpenAI's tts-1 ($30 for tts-1-hd), $30 on Deepgram Aura-2, $40 on ElevenLabs Flash and Turbo and $80 on ElevenLabs v3; OpenAI's gpt-4o-mini-tts and Gemini's TTS models bill per token instead. Pre-recorded transcription costs about $0.15 to $0.36 per audio hour: AssemblyAI Universal-2 $0.15 and Universal-3.5 Pro $0.21, ElevenLabs Scribe v2 $0.22, Deepgram Nova-3 $0.26, OpenAI gpt-transcribe $0.27 and Whisper $0.36. Streaming, speaker labels and idle connections cost extra.

Key takeaways:
- TTS per million characters: OpenAI tts-1 $15, tts-1-hd $30; Deepgram Aura-2 $30; ElevenLabs Flash/Turbo $40, v3 and Multilingual v2 $80.
- Pre-recorded transcription per audio hour: AssemblyAI Universal-2 $0.15, Universal-3.5 Pro $0.21, ElevenLabs Scribe v2 $0.22, Deepgram Nova-3 $0.26, OpenAI gpt-transcribe $0.27, Whisper $0.36.
- OpenAI's gpt-4o-mini-tts and Gemini's TTS models bill per token, so compare them on your own text, not on a per-character table.
- Streaming costs more and some vendors bill the whole session, including idle time; speaker labels are often an add-on.
- Voices are not portable between vendors, so a TTS fallback changes how your product sounds.

**What do speech APIs cost in 2026?** For **text-to-speech**, per million characters: OpenAI tts-1 $15 and tts-1-hd $30[1], Deepgram Aura-2 $30[2], ElevenLabs Flash and Turbo $40, ElevenLabs v3 and Multilingual v2 $80[3]. OpenAI's gpt-4o-mini-tts and Google's Gemini TTS models bill per token instead. For **pre-recorded transcription**, per audio hour: AssemblyAI Universal-2 $0.15 and Universal-3.5 Pro $0.21[4], ElevenLabs Scribe v2 $0.22[3], Deepgram Nova-3 about $0.26[2], OpenAI gpt-transcribe $0.27 and Whisper $0.36[1].

Those numbers are only the start. Streaming, speaker labels, idle sessions and the unit each vendor bills in change the real cost, and voice quality is something you can only judge by listening.

## Text-to-speech

| Model | Billing unit | Rate | Per 1M characters |
| --- | --- | --- | --- |
| OpenAI tts-1 | Characters | $15 per 1M | $15 |
| OpenAI tts-1-hd | Characters | $30 per 1M | $30 |
| Deepgram Aura-2 | Characters | $0.030 per 1K | $30 |
| ElevenLabs Flash / Turbo | Characters | $0.04 per 1K | $40 |
| ElevenLabs v3 / Multilingual v2 | Characters | $0.08 per 1K | $80 |
| OpenAI gpt-4o-mini-tts | Tokens | $0.60 per 1M text in, $12 per 1M audio out | Depends on audio length |
| Gemini 3.8 Flash TTS | Tokens | $0.50 per 1M text in, $9 per 1M audio out (until 31 Dec 2026) | Depends on audio length |

Table: From each vendor's pricing page on 2026-10-20.

Three notes from the pages:

- **ElevenLabs** keeps the per-unit rate the same across plans; plans change the included allowance, from 10,000 characters to over 12 million, with monthly fees from $6 to $990[3].
- **Gemini's TTS** rates double on 1 January 2027 for the 3.8 Flash TTS and Flash-Lite TTS models, to $1.00 per million text tokens and $18 and $12 per million audio tokens respectively, and both have a free tier[5].
- **Token-billed TTS** is priced on the audio produced, so the cost per character depends on how long the speech is. Measure it on your own scripts before comparing it with a per-character rate.

## Speech-to-text

| Model | Mode | Rate | Per audio hour |
| --- | --- | --- | --- |
| AssemblyAI Universal-2 | Pre-recorded | $0.15/hr | $0.15 |
| AssemblyAI Universal-3.5 Pro | Pre-recorded | $0.21/hr | $0.21 |
| ElevenLabs Scribe v2 | Pre-recorded | $0.22/hr | $0.22 |
| Deepgram Nova-3 (monolingual) | Pre-recorded | $0.0043/min | $0.26 |
| OpenAI gpt-transcribe | Pre-recorded | $0.0045/min | $0.27 |
| OpenAI Whisper | Pre-recorded | $0.006/min | $0.36 |
| OpenAI gpt-4o-mini-transcribe | Pre-recorded | est. $0.003/min | about $0.18 |
| AssemblyAI Universal-Streaming | Streaming | $0.15/hr | $0.15 |
| Deepgram Nova-3 (monolingual) | Streaming | $0.0048/min | $0.29 |
| Deepgram Flux English | Streaming | $0.0065/min | $0.39 |
| AssemblyAI Universal-3.6 Pro Realtime | Streaming | $0.45/hr base | $0.45 |
| OpenAI gpt-realtime-whisper | Streaming | $0.017/min | $1.02 |

Table: From each vendor's pricing page on 2026-10-20. OpenAI's gpt-4o-mini-transcribe is token-billed; the per-minute figure is OpenAI's own estimate.

What the table does not show:

- **Multilingual costs more** on several vendors: Deepgram Nova-3 multilingual is $0.31 per hour pre-recorded against $0.26 monolingual[2].
- **Speaker labels are often extra.** AssemblyAI adds $0.02 per hour for standard async speaker labels and $0.12 per hour on streaming[4].
- **Streaming can bill idle time.** AssemblyAI notes that streaming "is billed per session duration", so an open connection with nobody speaking still counts[4]. Close sessions when a call ends.
- **Promotional rates expire.** Deepgram marks its streaming prices as limited-time[2]; budget at the regular rate.

## Worked examples

> **Example: A thousand hours of recorded calls a month**
> Pre-recorded transcription, English, with speaker labels where they are an add-on.
> 
> - AssemblyAI Universal-3.5 Pro: 1,000 × ($0.21 + $0.02) = $230
> - ElevenLabs Scribe v2: 1,000 × $0.22 = $220
> - Deepgram Nova-3: 1,000 × $0.26 = $260
> - OpenAI gpt-transcribe: 1,000 × $0.27 = $270
> 
> The spread is small; accuracy on your callers' accents and audio quality will matter more than the rate.

> **Example: Ten million characters of narration a month**
> - OpenAI tts-1: $150
> - Deepgram Aura-2: $300
> - ElevenLabs Flash: $400
> - ElevenLabs v3: $800
> 
> Here the spread is fivefold, and so is the difference in voice range and expressiveness. Pick the tier per use: system notifications rarely need the most expressive voice; audiobook narration usually does.

## What to test before you choose

1. **Your audio, not a demo clip.** Phone-line audio, accents, crosstalk and domain vocabulary change word error rates more than vendor benchmarks show.
2. **Latency where it matters.** For voice agents, time to first audio byte (TTS) and time to final transcript (STT) decide whether a conversation feels natural.
3. **Languages.** Check every language you serve, including mixed-language speech.
4. **Features you need.** Diarization, timestamps per word, punctuation, profanity filtering, custom vocabulary — and which of them cost extra.
5. **Voice licensing.** Read the vendor's terms for commercial use of generated voices and for cloned voices.

## Reliability: why voice failover is different

For text models, failover to another provider serving the same model is close to invisible. For speech it is not:

- **TTS voices are vendor-specific.** A voice id on one vendor does not exist on another, so a fallback changes how your product sounds. Prefer failover to the same vendor on another account, or choose a fallback voice deliberately and accept the change.
- **Transcripts differ.** A fallback STT model produces different punctuation, casing and segmentation; downstream parsing must tolerate both.
- **Long files need chunking.** Split long recordings on silence and transcribe chunks in parallel, offsetting timestamps, rather than holding one connection open for an hour.

## Speech through INFRO

INFRO serves `POST /v1/audio/speech` and `POST /v1/audio/transcriptions` on your own provider connections, such as OpenAI, ElevenLabs and OpenAI-compatible speech services you connect. Speech requests are estimated per character of input and transcription per second of audio the provider reports, with the estimate in the `X-INFRO-Cost` header or `usage.cost`, so budgets at organization, project, member and key level cover voice traffic like any other. The text of a speech request passes your input guardrails. INFRO does not store synthesized audio, input text, recordings or transcripts from these endpoints; it records request metadata. Its documentation says plainly that `fallbacks` across TTS vendors rarely works, because voice ids are not portable.

## Summary

Convert every speech rate to the same unit before comparing: per million characters for TTS, per audio hour for transcription. Then add the costs a rate card hides — streaming, speaker labels, idle sessions, multilingual premiums, expiring promotions — and choose on your own audio, because accuracy and voice quality vary more than price does.

## Questions

### What is the cheapest speech-to-text API?

Among the published pre-recorded rates checked on 2026-10-20, AssemblyAI's Universal-2 at $0.15 per audio hour is the lowest, followed by Universal-3.5 Pro at $0.21 and ElevenLabs Scribe v2 at $0.22. Accuracy on your own audio, language coverage and add-ons such as speaker labels change the real cost, so test on your recordings.

### How much does the Whisper API cost?

OpenAI lists Whisper at $0.006 per minute, which is $0.36 per audio hour. Its newer gpt-transcribe model is listed at $0.0045 per minute, or $0.27 per hour, and gpt-4o-mini-transcribe at an estimated $0.003 per minute.

### How much does ElevenLabs text-to-speech cost per character?

ElevenLabs' API pricing page lists v3 and Multilingual v2 at $0.08 per 1,000 characters and Flash and Turbo at $0.04 per 1,000 characters. Plans change the included allowance, not the per-unit rate.

### Can I fail over between text-to-speech providers?

Technically yes, but voices are vendor-specific, so a fallback to another vendor changes the voice your users hear. For TTS, fail over to the same vendor on another account, or accept and design for a different fallback voice.

## Sources

1. Pricing — OpenAI. https://developers.openai.com/api/docs/pricing (accessed 2026-10-20)
2. Pricing — Deepgram. https://deepgram.com/pricing (accessed 2026-10-20)
3. API pricing — ElevenLabs. https://elevenlabs.io/pricing/api (accessed 2026-10-20)
4. Pricing — AssemblyAI. https://www.assemblyai.com/pricing (accessed 2026-10-20)
5. Gemini Developer API pricing — Google. https://ai.google.dev/gemini-api/docs/pricing (accessed 2026-10-20)