Google’s Gemini 3.5 Transcribe outperforms OpenAI’s GPT-Transcribe in multi-speaker accuracy
Google released Gemini 3.5 Transcribe on August 26, 2026, just four weeks after OpenAI’s GPT-Transcribe, creating a rare head-to-head comparison of two new transcription models. Both offer streaming and non-streaming versions, but Google’s model stands out for built-in multi-speaker diarization (up to three speakers) and word-level timestamps, eliminating the need for separate models. Google…
Key points
- Google’s Gemini 3.5 Transcribe cuts transcription time by 70% over its predecessor, Chirp 3, with built-in speaker labels and timestamps
- OpenAI’s GPT-Transcribe halves Whisper’s WER (from 40.37% to 19.27%) but requires extra models for speaker diarization and timestamps
- Google’s model supports **85+ languages** and integrates with other Gemini tools via function calling, while OpenAI’s focuses on cost efficiency
OpenAI’s model costs $0.0045 per minute for files and $0.017 for streaming, while Google’s pricing isn’t detailed in the article. The comparison highlights a clear trade-off: Google’s model excels in multi-speaker accuracy and timestamps, while OpenAI’s remains cheaper and simpler for single-speaker tasks or live captioning, where latency is critical.
The story so far
2 episodes →- Google’s Gemini 3.5 Transcribe outperforms OpenAI’s GPT-Transcribe in multi-speaker accuracythis story
Gemini 3.5 Transcribe vs OpenAI’s GPT-Transcribe
KDnuggets · 28 September 2026
Loading the full article…
This text was published by KDnuggets and written by Shittu Olumide. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Generative AI & Models
All →- OpenAI launches GPT-6 Sol and Luna with lower API prices and improved benchmarks · 13 src
- Anthropic unveils Claude Opus 5.5 with 40% lower costs and top safety ratings · 2 src
- Meta's Muse agent gives every user a full cloud computer running Ubuntu Linux · 9 src
- Fireworks AI releases Ember-1, a post-trained Kimi K3 model · 1 src
- Researchers test pseudo‑labeling to improve ASR on noisy police audio · 1 src
Comments
via GitHub Discussions