We use cookies. Cookies and browser storage keep you signed in and remember your preferences. With your permission we also measure how the site is used. You can any time. See our cookie policy.

New here? See how it works A 1-minute tour of the playground, models and your API key.

Chirp 3 vs MAI-Transcribe 2: API price, context & features · api14

Google: Chirp 3 vs Microsoft AI: MAI-Transcribe 2

A side-by-side comparison of Google: Chirp 3 ($0.06 / minute) and Microsoft AI: MAI-Transcribe 2 ($21.00 / minute) using current USD prices and capabilities from the api14 catalog. Both run through the same OpenAI-compatible API.

Specifications

	Google: Chirp 3	Microsoft AI: MAI-Transcribe 2
Provider	Google	Microsoft
Audio	$0.06 / minute	$21.00 / minute
Context window	-	-
Max output	-	-
Input	audio	audio
Output	transcription	transcription
Tool calling
Structured output
Reasoning control
Added	May 5, 2026	Sep 3, 2026

Key differences

Google: Chirp 3 is 100% cheaper on input tokens.

In depth

Google: Chirp 3 is a transcription model from Google, listed on api14 under the model id google/chirp-3. Microsoft AI: MAI-Transcribe 2 is a transcription model from Microsoft, listed under the model id microsoft/mai-transcribe-2. Both sit in the transcription category, so neither is a general chat or reasoning model. They exist to take audio in and return a transcription out. The two providers are direct competitors in cloud speech services, and api14 resells access to both behind one API surface.

On the facts available, the two models look very similar in shape. Both accept audio as their only input modality and both produce transcription as their only output modality. Neither lists a context length in tokens, so there is no documented difference in how much audio or text context each one handles. Neither lists supported parameters, so you should not expect either to expose tuning knobs such as temperature or prompt controls through this listing. There is no tool calling, no structured output mode and no reasoning mode recorded for either model, so those axes do not separate them. The one qualitative difference given is price on input: Chirp 3 is the cheaper side for input. Output pricing is recorded as a tie, meaning the two are about the same price on that side. So the practical cost gap, if any, comes from input, and it favors Google.

For picking between them, the decision is narrow because the capability sets match. If your workload is input heavy, meaning long audio files or high volumes of audio sent for transcription, Chirp 3 is the cheaper option on the input side and is the reasonable default. If your workload is dominated by output volume, the two are about the same price, so cost stops being a tiebreaker and you can choose on other grounds, such as which provider's transcription quality you have already validated on your own audio, or which vendor your organization already has a relationship with. Because neither model exposes tools, structured output or reasoning, do not plan around those features with either one. Treat both as pure speech to text endpoints and do your downstream parsing, formatting or extraction in your own code after the transcription comes back.

Calling either model through api14 works the same way, because the API is OpenAI compatible. You send your audio to the transcription endpoint with your api14 key, and the only thing that changes between the two providers is the model id string. Use google/chirp-3 to route to Google's model, or microsoft/mai-transcribe-2 to route to Microsoft's. The request shape, the authentication header and the response handling stay identical, so you can switch providers by changing one field, or run both side by side on the same audio to compare results before committing. This also means you can keep a single client library and a single set of retry and logging code while still having access to both vendors.

There is no case in the supplied facts where MAI-Transcribe 2 is clearly the better pick on a technical axis, since capabilities, context and modalities do not differ and output pricing is about the same. The honest position is that Chirp 3 has the edge on input price, and Microsoft's model is a reasonable alternative when you prefer that provider, when you have already tuned your pipeline around its output, or when you want to avoid depending on a single vendor. Since both are reachable through the same api14 endpoint with only the model id changing, keeping both in your configuration costs you very little.

Switch with one string

Both models use the same endpoint and request shape. Change the model value from "google/chirp-3" to "microsoft/mai-transcribe-2" and prompts, streaming and tools keep working.

Google: Chirp 3 details & playground →Microsoft AI: MAI-Transcribe 2 details & playground →

import fs from "node:fs";
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api14.com/api/v1",
  apiKey: process.env.API14_API_KEY,
});

const result = await client.audio.transcriptions.create({
  model: "microsoft/mai-transcribe-2",
  file: fs.createReadStream("audio.mp3"),
});
console.log(result.text);

Google: Chirp 3 vs Microsoft AI: MAI-Transcribe 2

Specifications

Key differences

In depth

Switch with one string

More comparisons