---
title: "Speech-to-text API cost calculator (per minute) | BroadRouter"
description: "Free speech-to-text cost calculator. Turn hours of audio, calls or voice notes into a monthly transcription bill and compare it with your current STT provider."
url: https://broadrouter.com/tools/speech-to-text-cost-calculator/
lang: en
updated: 2026-11-16
publisher: BroadRouter (a BroadNet site); API by BroadNet Inference, https://inference.broadnet.ai
---

# Speech-to-text cost calculator

Updated 2026-11-16

**In short:** Speech-to-text APIs bill per minute of audio: monthly cost = billed minutes × price per minute. Billed minutes depend on rounding (exact seconds, or each file rounded up to a minute) and on whether each stereo channel is billed. BroadNet starts at $0.002 a minute; Soniox lists less. Compare your volume against seven providers' list prices.

Free, no sign-up, runs in your browser: nothing you type is sent anywhere.

Audio volumeMeasure audio asHours of audio per monthPhone calls × average lengthVoice notes × average secondsAudio arrives asRecorded files (batch)Live streams (real-time)Hours per monthCalls per monthAverage call, minutesVoice notes per monthAverage note, secondsBilling incrementPer secondPer minute, each file rounded upChannels billed per file (2 for stereo billed per channel)LanguageArabic (including dialects)Multilingual / English and other languages

### Your current provider (optional)

NamePrice per minute

 Include published list prices of other providers

| Provider / model | Per minute | Billed minutes | Per month | Per year |
| --- | --- | --- | --- | --- |

[Test your audio with $5 free credit](https://inference.broadnet.ai/?utm_source=broadrouter&utm_medium=referral&utm_campaign=tools-speech-to-text-cost-calculator&utm_content=tool) ·

Estimates, before tax. BroadNet prices are its published per-audio-minute prices, synced 2026-11-14. Other providers' rows use the list prices in the table below, checked 2026-11-14, with each provider's stated channel rule; they ignore volume discounts, free tiers and add-ons such as diarization. Files on BroadNet go up to 25 MiB each.

[Report a wrong price or number](mailto:info@broadnet.ai?subject=Wrong%20price%20or%20number%20on%20Speech-to-text%20cost%20calculator&body=Page%3A%20https%3A%2F%2Fbroadrouter.com%2Ftools%2Fspeech-to-text-cost-calculator%2F%0A%0AWhat%20looks%20wrong%20(model%2C%20figure%20shown%2C%20correct%20figure%20and%20a%20source%20link%20if%20you%20have%20one)%3A%0A) (email to info@broadnet.ai, with this page's link and your inputs filled in).

## How it works

1.  Choose recorded files (batch) or live streams (real-time), then how you measure audio: total hours a month, phone calls with an average length, or voice notes with an average length in seconds.
2.  Choose the billing increment (per second, or each file rounded up to a whole minute) and how many channels are billed per recording.
3.  Choose Arabic or multilingual audio, and optionally type your current provider and its price per minute.
4.  Read the table: billed minutes and monthly cost for BroadNet's models, your provider and published list prices of seven other providers. Copy the link or download CSV.

## The formula

```
billed seconds per file = duration (per second)
                        or ceil(duration / 60) × 60 (per minute, rounded up)
billed minutes          = files × billed seconds per file / 60 × channels billed
monthly cost            = billed minutes × price per minute
```

Two settings change the bill more than most people expect. **Rounding:** a 20-second voice note billed per minute costs three times its length. **Channels:** a stereo call recording with one speaker per channel is two streams of audio, and providers that bill per channel charge for both. The "hours" mode treats your audio as one long total, so rounding does not apply to it. See [speech to text](https://broadrouter.com/glossary/speech-to-text/) in the glossary.

## Published speech-to-text list prices, compared

List prices per audio minute in USD for the cheapest standard pre-recorded (batch) option, plus the streaming (real-time) price, as published on each provider's own pricing page on 2026-11-14. BroadNet prices synced 2026-11-14.

| Provider and model | Per minute | Per hour | Streaming, per minute | Multi-channel audio | Notes and source |
| --- | --- | --- | --- | --- | --- |
| [BroadNet Transcribe Arabic](https://broadrouter.com/models/cohere-transcribe-arabic/) (BroadNet) | USD 0.0042 | USD 0.252 | No streaming price listed | Not stated | Arabic only, including dialects. [Pricing](https://inference.broadnet.ai/pricing?utm_source=broadrouter&utm_medium=referral&utm_campaign=tools-speech-to-text-cost-calculator&utm_content=compare-table) |
| [BroadNet Transcribe](https://broadrouter.com/models/cohere-transcribe/) (BroadNet) | USD 0.002 | USD 0.12 | No streaming price listed | Not stated | 14 languages, including Arabic. [Pricing](https://inference.broadnet.ai/pricing?utm_source=broadrouter&utm_medium=referral&utm_campaign=tools-speech-to-text-cost-calculator&utm_content=compare-table) |
| OpenAI gpt-transcribe | USD 0.0045 | USD 0.27 | USD 0.017 | Not stated | Listed as an estimated cost per minute (OpenAI re-checked 2026-11-15). whisper-1 and gpt-4o-transcribe (USD 0.006) and gpt-4o-mini-transcribe (USD 0.003) are marked deprecated and shut down on 2027-02-26. Streaming: gpt-live-transcribe, USD 0.017 a minute. [Official pricing](https://developers.openai.com/api/docs/pricing), checked 2026-11-14. |
| Google Cloud Speech-to-Text V2, standard | USD 0.016 | USD 0.96 | USD 0.016 | Each channel billed | First 500,000 minutes a month; lower tiers above that. Each request rounded up to the next second; each channel billed. Streaming: The page lists one standard Recognition price, USD 0.016 a minute for the first 500,000 minutes, with no separate streaming rate. [Official pricing](https://cloud.google.com/speech-to-text/pricing), checked 2026-11-14. |
| Amazon Transcribe, batch (US East) | USD 0.006 | USD 0.36 | USD 0.01 | Up to two channels included | USD 0.0001 per second in the AWS price list for us-east-1; streaming is USD 0.01 a minute. Up to two channels included; billed per second, no minimum. Streaming: Streaming, USD 0.01 a minute in us-east-1. [Official pricing](https://aws.amazon.com/transcribe/pricing/), checked 2026-11-14. |
| Azure AI Speech, batch (East US) | USD 0.003 | USD 0.18 | USD 0.016667 | Not stated | USD 0.18 per audio hour for batch; real-time is USD 1.00 per hour and fast transcription USD 0.36 per hour. Billed in one-second increments. Streaming: Real-time, USD 1.00 per audio hour. [Official pricing](https://azure.microsoft.com/en-us/pricing/details/cognitive-services/speech-services/), checked 2026-11-14. |
| Deepgram Nova-3, pre-recorded | USD 0.0043 to USD 0.0052 | USD 0.258 to USD 0.312 | USD 0.0048 to USD 0.0058 | Each channel billed | Monolingual (one language per request) USD 0.0043; multilingual USD 0.0052. Billed per second; all channels billed. Streaming: Nova-3 streaming, USD 0.0048 monolingual and USD 0.0058 multilingual, shown as promotional prices (regular USD 0.0077 and USD 0.0092). [Official pricing](https://deepgram.com/pricing), checked 2026-11-14. |
| Soniox, async (file) | USD 0.001667 | USD 0.1 | USD 0.002 | Not stated | USD 0.10 per audio hour for async file transcription and USD 0.12 per hour real-time, billed by audio and text tokens; 60+ languages at one price, Arabic listed as supported. Cheaper per hour than both BroadNet models. Billing increment and channel rule not stated. Streaming: Real-time, USD 0.12 per audio hour. [Official pricing](https://soniox.com/pricing), checked 2026-11-14. |
| AssemblyAI Universal-2, pre-recorded | USD 0.0025 | USD 0.15 | USD 0.0025 | Each channel billed | USD 0.15 per audio hour; Universal-3.5 Pro is USD 0.21 per hour. Multichannel audio billed per channel. Streaming: Universal-Streaming, USD 0.15 per hour, billed on WebSocket session time rather than audio sent. [Official pricing](https://www.assemblyai.com/pricing), checked 2026-11-14. |

List prices change and most providers discount at volume, so treat this as a starting point and check the linked page before you sign. Language support differs too: confirm that a provider transcribes your language and dialects, then test accuracy on your own recordings.

The full list, with billing increments, streaming prices and Arabic support for more than 20 products, is in our open [speech-to-text API price index](https://broadrouter.com/data/speech-to-text-api-prices/) (CSV and JSON).

## Worked example: a call centre in Arabic

5,000 calls a month averaging 4.5 minutes is 22,500 minutes of audio. Recorded in stereo, with the agent and the customer on separate channels, a provider that bills each channel counts 45,000 minutes. At USD 0.016 a minute that is USD 720 a month; mixed down to mono it is USD 360. On BroadNet Transcribe Arabic at USD 0.0042 a minute, 22,500 mono minutes cost USD 94.50.

## How to cut a transcription bill

-   Mix stereo to mono unless you need speakers separated by channel.
-   Trim silence and hold music before you upload; you pay for audio, not words.
-   Batch short clips only if your provider rounds per file; with per-second billing it makes no difference.
-   Use the single-language model when you know the language; multilingual pricing is often higher.

## Questions

### How much does speech-to-text cost per hour?

Multiply the per-minute price by 60. On BroadNet Inference, BroadNet Transcribe is $0.002 a minute ($0.12 an hour) and BroadNet Transcribe Arabic is $0.0042 a minute ($0.252 an hour). Published list prices for the cheapest batch option range from USD 0.10 to USD 0.96 an hour; Soniox lists the lowest, below BroadNet. See the table below.

### Is real-time speech-to-text more expensive than batch?

Usually. On the pricing pages we checked on 2026-11-14, streaming costs more than file transcription at Amazon Transcribe (USD 0.01 against USD 0.006 a minute), Azure (USD 1.00 against USD 0.18 an hour), Soniox and OpenAI, slightly more at Deepgram on its promotional price, and the same at Google Cloud and AssemblyAI. BroadNet lists file transcription only. Set "Audio arrives as" to live streams to compare.

### Are stereo call recordings billed twice?

Often, yes. Google Cloud, Deepgram and AssemblyAI bill every channel, so a 10-minute stereo call is 20 billed minutes. Amazon Transcribe includes up to two channels in its price. If you do not need speakers on separate channels, mix to mono before you upload and the bill halves on per-channel providers.

### What is the difference between per-second and per-minute billing?

With per-second billing you pay for the exact duration. With per-minute billing each file is rounded up to the next whole minute, so a 20-second voice note is billed as a full minute: three times its length. For short clips, the billing increment matters more than the list price. Set "Billing increment" to see the difference.

### How much does it cost to transcribe WhatsApp voice notes?

Choose "Voice notes", enter how many you receive a month and their average length. At $0.0042 a minute, 10,000 Arabic voice notes of 30 seconds cost about $21 billed per second. [How to transcribe WhatsApp voice notes](https://broadrouter.com/guides/transcribe-whatsapp-voice-notes/).

### Is there a free speech-to-text API credit?

BroadNet Inference gives a $5 starting credit, and both transcription models run on it. That is about 1,190 minutes of Arabic audio or 2,500 minutes on the multilingual model, enough to test your own recordings.

### Which speech-to-text API is best for Arabic?

Price is only half the answer: dialect accuracy varies a lot between providers. BroadNet reported its Arabic model #2 of 37 on the Open Universal Arabic ASR Leaderboard in September 2026; the live board listed it #4 of 39 on 15 November 2026. Test 50 to 100 of your own recordings and compare [word error rate](https://broadrouter.com/glossary/word-error-rate/). [Arabic speech-to-text buyer's guide](https://broadrouter.com/blog/arabic-speech-to-text-api-buyers-guide/).

## Go deeper

-   [Arabic speech-to-text API: quick start](https://broadrouter.com/guides/arabic-speech-to-text-api/)
-   [Transcribe WhatsApp voice notes](https://broadrouter.com/guides/transcribe-whatsapp-voice-notes/)
-   [Arabic speech-to-text API buyer's guide](https://broadrouter.com/blog/arabic-speech-to-text-api-buyers-guide/)
-   [BroadNet Transcribe Arabic: model page](https://broadrouter.com/models/cohere-transcribe-arabic/)
-   [BroadNet Inference prices](https://broadrouter.com/pricing/)

## More free tools

-   [**LLM API cost calculator**Price a workload on every model at once, with cache hits, reasoning and a flat-plan check.](https://broadrouter.com/tools/llm-api-cost-calculator/)
-   [**LLM token counter**Exact token counts with each model's real tokenizer, with a token view and cost.](https://broadrouter.com/tools/token-counter/)
-   [**Arabic token counter**Arabic vs English token counts per tokenizer: the "Arabic tax" in tokens and money.](https://broadrouter.com/tools/arabic-token-counter/)
-   [**Tokens to words converter**Tokens ↔ words ↔ characters ↔ pages, by language and content type.](https://broadrouter.com/tools/tokens-to-words/)
-   [**Context window calculator**Will it fit? Window used per model, room left for output, cost to send.](https://broadrouter.com/tools/context-window-calculator/)
-   [**Flat-rate vs per-token LLM cost calculator**Break-even between an unlimited plan and paying per token, per agent.](https://broadrouter.com/tools/flat-rate-vs-per-token-calculator/)
-   [**Prompt caching savings calculator**Bill with and without caching, by prefix share and hit rate.](https://broadrouter.com/tools/prompt-caching-calculator/)
-   [**Image token calculator**Tokens and cost per image under each model's documented rule. Drop an image; nothing is uploaded.](https://broadrouter.com/tools/image-token-calculator/)
-   [**Self-hosting vs API cost calculator**GPU rental against your current API and BroadNet per-token and flat-plan prices: cost per 1M tokens and break-even volume.](https://broadrouter.com/tools/self-hosting-vs-api-calculator/)
-   [**Embedding cost calculator**Index, update and query costs for a corpus at OpenAI, Google, Voyage, Mistral and Jina list prices.](https://broadrouter.com/tools/embedding-cost-calculator/)
-   [**LLM rate limit and concurrency calculator**Which limit binds first: RPM, TPM or concurrency, with Little's law.](https://broadrouter.com/tools/llm-rate-limit-calculator/)
-   [**JSON schema generator for LLM structured output**Sample JSON to strict schema, Pydantic, Zod and a ready request.](https://broadrouter.com/tools/json-schema-generator/)
-   [**OpenAI-compatible API request builder**One request, copied as curl, Python, Node or the OpenAI SDKs.](https://broadrouter.com/tools/api-request-builder/)
-   [**Audio file splitter for speech-to-text limits**Chunk count, overlap and copy-ready ffmpeg commands to fit a transcription file-size limit.](https://broadrouter.com/tools/audio-splitter-ffmpeg/)
-   [**Agent run cost estimator**Context growth per step, total tokens and cost of one agent run, with a DeepSeek session comparison.](https://broadrouter.com/tools/agent-run-cost-estimator/)
-   [**OpenAI to BroadNet migration helper**Paste OpenAI SDK code; get base URL, key and model edits, a diff and the checks that remain.](https://broadrouter.com/tools/openai-to-broadnet-migration/)
-   [**Arabic text normalizer**Tashkeel, tatweel, alef, ya, ta marbuta, digits and invisible characters, with a matching Python function.](https://broadrouter.com/tools/arabic-text-normalizer/)
-   [**RAG chunking calculator**Chunks, embedding tokens, input per query, context check and monthly LLM cost for a RAG pipeline.](https://broadrouter.com/tools/rag-chunking-calculator/)
-   [**max\_tokens calculator for reasoning models**Answer length + reasoning budget + margin = a max\_tokens that does not cut the answer off.](https://broadrouter.com/tools/reasoning-token-budget/)
-   [**LLM batch job planner**Run time, binding limit, cost per model and the flat sessions needed to finish a big job by a deadline.](https://broadrouter.com/tools/llm-batch-job-planner/)
-   [**LLM JSON output validator**Parse errors by line and column, safe repairs, cut-off detection and JSON Schema checks for model output.](https://broadrouter.com/tools/llm-json-output-validator/)
-   [**Prompt token diff**Two prompt versions side by side: changed lines, exact token delta and the monthly cost difference.](https://broadrouter.com/tools/prompt-token-diff/)

Try it on your real workload. New accounts start with $5 of credit for Gemma and transcription. DeepSeek, GLM and Qwen3.8 need a top-up.

[Get started](https://inference.broadnet.ai/account/?utm_source=broadrouter&utm_medium=referral&utm_campaign=tools-speech-to-text-cost-calculator&utm_content=page-cta) ·

**Sources**

-   [BroadNet Inference machine-readable prices (pricing.json)](https://inference.broadnet.ai/pricing.json?utm_source=broadrouter&utm_medium=referral&utm_campaign=tools-speech-to-text-cost-calculator&utm_content=sources)
-   [BroadNet Inference API reference (transcription limits)](https://inference.broadnet.ai/docs?utm_source=broadrouter&utm_medium=referral&utm_campaign=tools-speech-to-text-cost-calculator&utm_content=sources)
-   [OpenAI gpt-transcribe pricing page, checked 2026-11-14](https://developers.openai.com/api/docs/pricing)
-   [Google Cloud Speech-to-Text V2 pricing page, checked 2026-11-14](https://cloud.google.com/speech-to-text/pricing)
-   [Amazon Transcribe pricing page, checked 2026-11-14](https://aws.amazon.com/transcribe/pricing/)
-   [Azure AI Speech pricing page, checked 2026-11-14](https://azure.microsoft.com/en-us/pricing/details/cognitive-services/speech-services/)
-   [Deepgram Nova-3 pricing page, checked 2026-11-14](https://deepgram.com/pricing)
-   [Soniox pricing page, checked 2026-11-14](https://soniox.com/pricing)
-   [AssemblyAI Universal-2 pricing page, checked 2026-11-14](https://www.assemblyai.com/pricing)
-   [Amazon Transcribe price list API, us-east-1, checked 2026-11-14](https://pricing.us-east-1.amazonaws.com/offers/v1.0/aws/transcribe/current/us-east-1/index.json)
-   [Azure Retail Prices API, Azure Speech S1, East US, checked 2026-11-14](https://prices.azure.com/api/retail/prices)
