AEO for speech-to-text APIs: AI search visibility ranking
How AI search engines rank speech-to-text APIs by visibility and citations. 20 brands measured quarterly across Google AI Mode: which brands the AI names in answers, which domains it cites as sources, and how the leaders compare. Speech-to-text APIs used to transcribe calls, meetings, media, and voice interactions into searchable text and downstream workflows. Composite score: 70% visibility (% of AI answers naming the brand) + 30% citation rate (% citing the brand's domain). Full methodology →
Download this AEO ranking as a PDF
We'll email it to you. One-off send — no list, no follow-up, no surprise marketing.
When buyers inquire about Speech-to-text APIs, AI engines most frequently mention AssemblyAI as the leading choice, with a visibility of 62.5 percent. OpenAI and Deepgram also receive notable recognition, showcasing their robust presence in the market. AssemblyAI's prominence highlights its significant influence among these influential technologies.
The rankings reflect specific patterns based on citation sources. AI engines primarily access information from well-regarded platforms such as google.com, youtube.com, and reddit.com. The presence of aggregator and review sites suggests that comprehensive user feedback and community discussions play important roles in shaping these rankings.
For buyers, this list emphasizes the importance of examining brands with strong AI visibility. Consider recent reviews, third-party validation, and each brand's topical authority within the Speech-to-text API subcategory. Focus on those that are frequently validated by trusted sources or have consistent positive feedback from AI engines.
At a glance
What we observed in this category
AssemblyAI leads the Speech-to-text APIs category with a composite score of 51.2, ahead of OpenAI and Deepgram who both sit at 43.8. The gap matters because AssemblyAI is the only brand in the top three converting visibility into citations, recording a 25.0% citation rate while OpenAI and Deepgram sit at 0.0% despite matching its 62.5% visibility. All other ranked brands score below 13 on composite, meaning the top three hold a structural advantage that the remaining 17 brands are not close to closing.
A sharp divergence between visibility and citation runs through this category. OpenAI and Deepgram are named in AI answers at the same rate as AssemblyAI yet receive zero citations, making them recognised but not trusted as sources. Conversely, Speechmatics, Gladia, and Soniox each hold a 12.5% citation rate with zero visibility, meaning the AI references their content without surfacing their brand names in responses. This split shows that being mentioned and being cited are rewarded by different content signals in this category.
Google AI Mode is the dominant engine across all 10 ranked brands, listed as the top engine for every brand in the dataset. The top cited sources include google.com, youtube.com, and reddit.com alongside niche properties such as futureagi.substack.com, llmapi.ai, and voicewriter.io. The presence of Substack and small AI-focused domains alongside platform giants suggests Google AI Mode is drawing on a mix of community commentary and specialist review content when constructing answers about speech-to-text APIs.
Movers & shakers since last refresh
Biggest visibility risers
-
OpenAI 12% → 62% · rank #8 → #2+50pp
-
Nuance 0% → 12% · rank #18 → #6+12pp
Biggest visibility fallers
-
Speechmatics 25% → 0% · rank #5 → #7-25pp
-
Gladia 25% → 0% · rank #4 → #8-25pp
-
Amazon Transcribe 12% → 0% · rank #7 → #10-12pp
The AEO ranking
| # | Brand | Visibility | Citation | Top engine |
|---|---|---|---|---|
| 1 |
assemblyai.com
|
62% | 25% | Google AI Mode |
AssemblyAI is the only top-three brand converting visibility into citations, with a 25.0% citation rate against a category average of 3.8%, giving it a composite score 7.4 points above OpenAI. |
||||
| 2 |
openai.com
|
62% | 0% | Google AI Mode |
OpenAI jumped from rank 8 to rank 2 with a 50.0 percentage point visibility gain, yet its citation rate remains 0.0%, leaving it named but uncited across all sampled queries. |
||||
| 3 |
deepgram.com
|
62% | 0% | Google AI Mode |
Deepgram matches OpenAI on visibility (62.5%) and composite score (43.8%) but also records 0.0% citations, placing it in the same named-but-not-trusted position as its co-ranked peer. |
||||
| 4 |
cloud.google.com
|
12% | 12% | Google AI Mode |
Google Cloud sits well below the top three at 12.5% visibility, but uniquely balances that with a matching 12.5% citation rate, making it the only brand outside the top three with a citation above zero. |
||||
| 5 |
openai.com
|
12% | 0% | Google AI Mode |
Whisper API shares the openai.com domain with OpenAI yet scores only 8.8 composite against OpenAI's 43.8, indicating the API sub-product receives far less AI answer coverage than the parent brand. |
||||
| 6 |
nuance.com
|
12% | 0% | Google AI Mode |
| 7 |
speechmatics.com
|
0% | 12% | Google AI Mode |
| 8 |
gladia.io
|
0% | 12% | Google AI Mode |
| 9 |
soniox.com
|
0% | 12% | Google AI Mode |
| 10 |
aws.amazon.com
|
0% | 0% | Google AI Mode |
| 11 |
microsoft.com
|
0% | 0% | Google AI Mode |
| 12 |
rev.ai
|
0% | 0% | Google AI Mode |
| 13 |
symbl.ai
|
0% | 0% | Google AI Mode |
| 14 |
fathom.video
|
0% | 0% | Google AI Mode |
| 15 |
happyscribe.com
|
0% | 0% | Google AI Mode |
| 16 |
tiro.is
|
0% | 0% | Google AI Mode |
| 17 |
verbit.ai
|
0% | 0% | Google AI Mode |
| 18 |
otter.ai
|
0% | 0% | Google AI Mode |
| 19 |
fireflies.ai
|
0% | 0% | Google AI Mode |
| 20 |
alphacephei.com
|
0% | 0% | Google AI Mode |
Sources AI engines trust in this category
Across the 8 buyer-intent queries we ran on speech-to-text APIs, these are the domains Google AI Mode cited most often. If you're not on this list — or if your competitors are — that's a concrete PR / linkbuilding target.
How to read this AEO ranking
Four things worth knowing before you act on the numbers above. These are the same definitions across every industry page — for category-specific observations, see the What we observed section above (where available) and the per-brand insights inline in the ranking.
Visibility = being named
A brand's visibility % is the share of AI answers that mention it by name in the response prose. This is who AI engines actively recommend to the buyer. More on visibility →
Citation rate = being trusted
Citation rate is the share of AI answers that include the brand's domain as a clickable source link. This is what the AI treats as authoritative evidence, different from being named. More on citation rate →
Top engine differs by brand
The "top engine" column shows which AI surface each brand performs best on. Big gaps between a brand's score across engines usually points to specific content or schema gaps. How AI engines pick sources →
AEO rankings move month to month
AI engines re-crawl and re-rank on shorter cycles than classical search. We re-audit every brand on this list at least every 90 days and refresh this page automatically. How AI search ranking works →
Run an AEO audit on your own speech-to-text APIs brand
The brands above were curated from public market-leader lists. Want the same answer engine optimisation measurement against your own brand — including the queries you appear on, which competitors get named instead, and a prioritised fix list? Run a free preview.
Frequently asked about AEO for speech-to-text APIs
Who leads AEO in Speech-to-text APIs?
AssemblyAI leads with a composite score of 51.2 and is the only top-three brand with a citation rate above zero, recording 25.0% against a category average of 3.8%.
What sources does Google AI Mode cite most for Speech-to-text APIs?
The top cited sources in this category are google.com, youtube.com, and reddit.com, followed by niche properties including futureagi.substack.com, llmapi.ai, and voicewriter.io.
Which brands have high visibility but no citations in this category?
OpenAI and Deepgram both reach 62.5% visibility but record 0.0% citation rate, meaning AI answers name them without drawing on their content as sources.
Are any brands cited by AI without appearing in its answers?
Yes. Speechmatics, Gladia, and Soniox each hold a 12.5% citation rate despite 0.0% visibility, meaning their content is used as a source but their brand names are not surfaced in responses.
Which brands have lost the most AEO ground in Speech-to-text APIs recently?
Gladia fell from rank 4 to rank 8 with a 25.0 percentage point visibility drop and a 50.0 percentage point citation drop, while Speechmatics fell from rank 5 to rank 7 with identical visibility and citation losses.
How concentrated is AEO performance in the Speech-to-text APIs category?
The top three brands (AssemblyAI, OpenAI, Deepgram) each hold 62.5% visibility against a category average of 11.2%, and all brands ranked 4 and below score 12.5% visibility or less.