Speechace API

Number of APIs: 13

Speechace is a Speech Recognition API for fluency and pronunciation assessment. Our patented technology is unique in its ability to score a learner's speech and pinpoint individual syllable and phoneme level mistakes in a user's pronunciation in real time.

NEW! Task Achievement and full PTE Speaking assessment

Introducing the Score Task API with support for Describe-Image, Retell-Lecture, and Answer-Question content scoring

The new Score Task API now supports advanced evaluation of the content of a spontaneous response, and how well it accomplished the task set out in the question. The API supports scoring tasks with either audio or text only and provides a content score aligned with the PTE content score scale.

NEW! French and Spanish Scripted and Spontaneous Speech Assessment with Task Achievment

Speechace Pronunciation Assessment API and Fluency Assessment API now support French and Spanish

Speechace API v9.9 now supports assessment of pronunciation and fluency in 4 new dialects: (fr-fr, fr-ca, es-es, es-mx):

Pronunciation assessment with utterance, word, syllable, and phoneme level scores
Fluency assessment with utterance and segment level scores
Scores provided on 0..100 scale and CEFR
Available today in all regions
Premium API support for Spontaneous Speech transcription + Scoring in French and Spanish
Premium API support for Task Achievment scoring in French and Spanish

See the API examples for details:
Score Pronunciation API

Score Fluency API

Transcribe & Score API

Score Task API

API SKUs and Use Cases

The Speechace API comes in 3 SKUs and can be used to assess and score a variety of speaking activities.

API SKU	Use Cases
BASIC	Pronunciation Assessment API Word and Sentence speaking activities Multiple Choice speaking activities
PRO	Fluency Assessment API Read Aloud, Repeat Sentence IELTS/PTE Fluency practice activities
PREMIUM	Spontaneous Speech Assessment API CEFR, IELTS, TOEIC, PTE Speaking Proficiency Assessment Free speaking activities Relevance Assessment Comprehensive skill assessment: Grammar, Vocabulary, Coherence, Pronunciation, Fluency Task Achievement

API Keys

You need a Speechace API Subscription to obtain a key. You can obtain a Speechace key at https://www.speechace.com/#api-plans

API Regions

The Speechace API currently supports the following regions. Your API Key is tied to a specific region. You can request additional API Keys under the same Subscription within other regions.

Region	API Endpoint
US West (Oregon)	https://api.speechace.co
AP Southeast (Singapore)	https://api2.speechace.com
EU West (Ireland)	https://api4.speechace.com
AP South (Mumbai)	https://api5.speechace.com

API Versioning

The Speechace API versioning is explicit in the request URL.

Up to v9.0 Speechace API versioning was explicit with the caller specifying the exact version they use in the request url. From v9.0 onwards the API caller can choose to take silent minor version updates.

Request url	Resulting version
https://api.speechace.co/scoring/text/v9/json?	This request will use the latest available minor version of v9 (e.g. v9.0 or v9.1 or v9.2 etc.)
https://api.speechace.co/scoring/text/v9.0/json?	This request will explicitly use the minor version v9.0

Major version history:

version	url	notes
1	https://api.speechace.co/api/scoring/text/v0.1/json?	v1
2	https://api.speechace.co/api/scoring/text/v0.2/json?	v2
3	https://api.speechace.co/api/scoring/text/v0.3/json?	v3
5	https://api.speechace.co/api/scoring/text/v0.5/json?	v5
7	https://api.speechace.co/api/scoring/text/v0.7/json?	v7
9	https://api.speechace.co/api/scoring/text/v9/json?	current latest production major version
9.2	https://api.speechace.co/api/scoring/text/v9.2/json?	New minor version scoring models.
9.4	https://api.speechace.co/api/scoring/text/v9.4/json?	New minor version Transcription models
9.5	https://api.speechace.co/api/scoring/text/v9.5/json?	New noise reduction enhacement. Beta Grammar, Coherence, Vocab feedback metrices. Current latest production minor version
9.7	https://api.speechace.co/api/scoring/text/v9.7/json?	Noise and aligment updates. New relevance scoring model Markup language support for letter to phoneme mapping
9.8	https://api.speechace.co/api/scoring/text/v9.8/json?	Pronunciation model update Grammatical accuracy model update
9.9	https://api.speechace.co/api/scoring/text/v9.9/json?	Core model updates Major performance enhancement

You can keep up with new version announcements by subscribing to the Speechace Blog at https://www.speechace.com/blog/

Samples

You can download samples showing both:

frontend (recording audio, passing to backend, showing scores)
backend (calling speechace API, passing scores to frontend)

The samples are available at https://github.com/speechace/speechace-api-samples

API Limits

The Speechace API has the following limits per the API SKU license: Basic, Pro, or Premium.

Limit	API SKU (BASIC)	API SKU (PRO)	API SKU (PREMIUM)
Max Audio Length	15 seconds	45 seconds	2 minutes
Max File Size	1.9MB	2.5MB	3.8MB

Recording Audio

Speechace supports audio in most formats used on the web (e.g. wav, mp3, m4a, ogg, webm, aiff). We strongly recommend recording at the following settings to optimize your file size and increase performance:

sample size: 16-bit
sample rate: 16Khz
channels: 1 (i.e. mono)

Terms of Service

By using the Speechace API you agree to the Privacy Policy and the Terms of Service.