---
title: "Google Cloud STT v2"
type: "AI Tool"
url: "https://aidemos.com/tools/google-cloud-stt-v2"
description: "We ran Google Cloud STT v2 on three batch transcriptions and got word-level timestamps; overlap and code-switching were uneven, with no speaker labels."
category: "audio-speech"
website: "https://cloud.google.com/speech-to-text/v2/docs"
published: "2026-08-24T00:52:41.334362+00:00"
updated: "2026-08-24T02:13:05.185036+00:00"
---

# Google Cloud STT v2

Batch transcription with word timestamps, but uneven accuracy on overlap and code-switching.

## TL;DR Verdict

**Useful batch STT, but not a clear winner on hard audio**

**Where it wins:**
- You need batch speech-to-text with word-level timestamps for long audio.
- You are comparing engines on technical narration and can tolerate uneven performance on harder clips.
- You want a straightforward API-backed transcription workflow with published Standard-tier pricing.

**Main limitation:** You need speaker diarization or confidence scores in the published chirp configuration.

**Pricing:** Free tier $0 · V2 Standard — batch or real-time $0.016/min ($0.96/audio-hour) · V2 Dynamic Batch $0.004/min ($0.24/audio-hour) · Volume discounts as low as ~$0.004/min

`Batch STT` · `Word timestamps` · `Hard audio tested` · `Standard tier`

**Website:** [Visit Google Cloud STT v2](https://cloud.google.com/speech-to-text/v2/docs)

> **Useful batch STT, but not a clear winner on hard audio**
>
> Google Cloud STT v2 completed all three benchmarked batch transcriptions and returned word-level timestamps, but performance was uneven: it was strongest on the medical-jargon file and much weaker on overlapping speech and Spanish-English code-switching. The published Standard-tier run cost about $1.39 across the three inputs, and the output lacked speaker labels and confidence scores in this configuration.

## Demo Recording

[Video: Google Cloud STT v2 demo recording](https://cdn.futuresmart.ai/public/aidemos/e77cf66cfb3e43f492745110a5ccdc8b.mov?v=1)
*Video — Terminal screen recording of the benchmark run selecting Google Cloud STT v2 and showing the three scored inputs and their run summaries.*

## Feature-by-Feature Breakdown

### Batch Speech Transcription

**Verdict:** Works as a batch transcription engine, but accuracy varied a lot by audio type.

Google Cloud STT v2 can take long-form audio and return a complete transcript in batch. The benchmark exercised it on overlapping crosstalk, technical medical narration, and bilingual code-switching, showing the same transcription capability across varied audio inputs.

**Input:**

> **Audio**

**Output:**

**Input:**

> **Audio**

**Output:**

**Input:**

> **Audio**

**Output:**

**Bottom line:** It reliably produced transcripts, but the quality spread was wide: acceptable on the technical narration, weak on crosstalk, and very weak on code-switching.

### Word-Level Timestamping

**Verdict:** Word offsets are present in the raw response for all three runs.

The API response can include per-word start and end offsets alongside the transcript. In the benchmark runs, these timestamps were present consistently, providing timing metadata for captions, alignment, and search indexing.

**Input:**

> **Audio**

**Output:**

**Input:**

> **Audio**

**Output:**

**Input:**

> **Audio**

**Output:**

**Bottom line:** Word timing is available consistently, but the output metadata is sparse beyond that.

## Official pricing

The benchmark used the V2 Standard tier; Dynamic Batch is cheaper but was not used for the published run.

| Plan | Price | Notes |
| --- | --- | --- |
| Free tier | $0 | 60 audio-minutes per month; applies across V1 and V2. |
| V2 Standard — batch or real-time ★ (tested) | $0.016/min ($0.96/audio-hour) | Pay-as-you-go; this is the tier the benchmark was billed at. |
| V2 Dynamic Batch | $0.004/min ($0.24/audio-hour) | Lower-urgency processing with no latency guarantee; not used in the published benchmark. |
| Volume discounts | as low as ~$0.004/min | Large monthly commitments; requires sales contact. |
| V1 API | $0.016/min | Same headline rate as Standard. |

*Free tier is 60 audio-minutes per month. The benchmark used about 86 minutes across three inputs, so one full pass exceeds the free allowance.*

## Is It Right For You?

**Use it if**
- You need batch speech-to-text with word-level timestamps for long audio.
- You are comparing engines on technical narration and can tolerate uneven performance on harder clips.
- You want a straightforward API-backed transcription workflow with published Standard-tier pricing.

**Skip it if**
- You need speaker diarization or confidence scores in the published chirp configuration.
- You need strong multilingual or code-switching accuracy.
- You need streaming latency numbers from this benchmark, because streaming was not measured here.

## Classification

- **Category:** audio-speech
- **Subcategory:** audio-to-text
- **Type:** speech
- **Built for:** Other

## Frequently Asked Questions

**Q: Does Google Cloud STT v2 return speaker labels in this benchmark?**

No. The published chirp run returned no speaker labels, and the diarization request was rejected as unsupported for chirp in the report. The notes say chirp_3 is the model to re-run if diarization becomes available.

**Q: Does it include word-level timestamps?**

Yes. The raw JSON responses show per-word startOffset and endOffset fields on all three inputs.

**Q: How accurate was it on technical jargon?**

That was its best input: 9.42% WER on the medical-jargon file, with 2,746 words returned against a 2,728-word reference. It still missed terms such as diaphysis, medullary, sesamoid, and trabeculae.

**Q: How did it do on overlapping speech?**

It struggled on the crosstalk file: 35.89% WER, with 2,183 deletions against a 7,579-word reference.

**Q: How did it do on bilingual code-switching?**

Poorly. The benchmark recorded 41.72% WER and only 3.8% recall on Spanish token types, so it handled the Spanish portion very weakly in this setup.

**Q: What does it cost?**

The benchmark used the V2 Standard tier at $0.016/min, or $0.96 per audio-hour. The vendor also lists a 60-minute free tier and a cheaper Dynamic Batch option at $0.004/min, but those were not the figures used for the published run.

**Q: Was streaming latency measured?**

No. This was a batch benchmark, so streaming latency was not measured.

## Similar Tools

AI tools similar to Google Cloud STT v2:

- [Gladia](https://aidemos.com/tools/gladia) — Batch STT with rich JSON metadata and strong jargon recall, but weak overlap handling and a channel-duplication caveat on bilingual audio.
- [Speechmatics](https://aidemos.com/tools/speechmatics) — Strong batch STT for hard English audio, but weak on code-switching as configured.
- [OpenAI](https://aidemos.com/tools/openai) — Batch speech-to-text with word timestamps, but a strict upload cap and weak multilingual performance make it a mixed fit for hard audio.
- [AWS Transcribe](https://aidemos.com/tools/aws-transcribe) — Batch speech-to-text with timestamps and speaker labels, but weak on multilingual audio.
- [ElevenLabs](https://aidemos.com/tools/elevenlabs) — Natural-sounding voice cloning and narration, but with only approximate voice identity.
- [Rev AI](https://aidemos.com/tools/rev-ai) — Low-cost batch speech-to-text with structured JSON, word timestamps, and speaker labels, but mixed accuracy on hard audio and weak multilingual recall.
- [GroqCloud Whisper Large-v3](https://aidemos.com/tools/groqcloud-whisper-large-v3) — Fast batch transcription for under-cap audio, with strong jargon handling but a hard 25 MB ceiling.

## Need a custom AI solution for this use case?

If you are looking to build a custom speech-to-text transcription, batch audio transcription, or captioning workflow for your business or internal workflow, email us at [contact@futuresmart.ai](mailto:contact@futuresmart.ai).

### Found something inaccurate or missing?

We try to keep our AI research accurate and useful. If you found outdated information, an issue, or have a suggestion, email us at [collaborate@aidemos.com](mailto:collaborate@aidemos.com).