Skip to content

Lecture transcription and captions for colleges, schools and online courses

Turn every class into searchable captions for $0.001/min, then turn the notes into narrated audio lessons with text to speech.

4.76% word error rate on 8 kHz phone, accented and noisy audioLive or recorded classes$0.001/min
Voice the notice that goes with your transcriptsSpeak example
Hi Priya, this is Westbrook College. The transcript for Biology 101 on Thursday at 9:00 AM is ready in your course page.
120 characters

Preview a lecture transcript notice

Lecture transcription turns the spoken audio of a class into written text that students can read, search and caption. Upload a recording or stream a live class to PyAI Hear, the speech to text model, and get lecture to text output with timestamps for $0.001/min, so a full semester of class transcription for a course costs less than a textbook. Hear sits on the same voice infrastructure as Speak, so the transcript of a class and the narrated version of its notes come from one API account. PyAI does not record the classroom itself, so the audio comes from your lecture capture system or a mic you already use.

How lecture transcription works

Once the lecture capture system is connected, every class is transcribed on its own. Teaching staff only check terms and publish.

1Class recorded

The room system or a clip on mic records the lecture.

2Audio sent

The file or live stream goes to PyAI Hear.

3Text returned

A timed transcript comes back for the class.

Recorded class

CaptionsTimed captions added to the video

SummaryRecap writes a short class summary

Live class

StreamingText appears as the lecturer speaks

SavedThe full transcript is stored after class

4Published

Students search, read and review the class on the course page.

Lecture transcription use cases

Colleges, schools and course creators use lecture transcription wherever spoken teaching needs to be read later. Each use below starts from the same class audio.

Captions

Live and recorded captions

Add captions to lecture videos and live streams.

Captions help students who are Deaf or hard of hearing and anyone watching in a noisy place. Recorded classes get timed captions, and live classes stream text as the lecturer speaks.

Study

Searchable class transcripts

Give students a transcript they can search before an exam.

A student types a term and jumps to the minute it was explained. Transcripts also fill the gaps in notes from a fast lecture.

Lecture to text

Notes and summaries from recordings

Turn long recordings into clean text and short summaries.

Split long sessions into parts of 30 to 45 minutes for easier review. Recap can write a short summary of each class for the course page.

Research

Seminars, interviews and office hours

Transcribe seminars, guest talks and research interviews.

Researchers get speaker separated text from interviews and focus groups, ready to quote and code without hours of typing.

Course audio

E learning voice over

Voice course modules from a script instead of a studio.

Text to speech for elearning turns a written lesson into narration with PyAI Speak for $0.04/min. Edit the script and regenerate the clip when a slide changes.

Languages

Transcripts for other languages

Help international students follow along.

Hear transcribes many languages, so a class in one language can be read and translated by students who are still learning it.

Lecture transcription cost calculator

Set the recordings a month, the audio length and how long a recording takes your staff by hand. The total covers transcription.

Estimated monthly cost$2.00a month for 2,000 recordings, 2,000 min

Transcription (Hear)$2.00

133 hstaff typing time saved each month

Lecture speech to text at scale

Lecture audio is hard: one speaker far from the mic, students asking questions across the room and long sessions with few breaks. PyAI Hear handles long files and live streams for $0.001/min, so a whole course library can be transcribed without a budget fight. Transcripts come back with timestamps for captions, and Recap can write a short summary of each class. For course narration, PyAI Speak returns WAV or PCM audio with the first byte in about 53 ms.

4.76%word error rate on 8 kHz phone, accented and noisy audio
1.59%word error rate on clean audio
~200 msto the first live transcript
92 to 247xreal time for batch files
Ask before recordingGet the lecturer's permission and tell students when a class is recorded and transcribed.
PII checks with TraceTrace can flag names and personal details in transcripts before they are shared beyond the class.
Keep access to the classShare transcripts in the course page for enrolled students rather than on open links.
Encrypted call recordsRecordings, transcripts and knowledge files sit in encrypted storage scoped to your organization, so no other workspace can read your calls.

Student privacy for class transcription

Class recordings hold the voices of students and staff, so handle them with care. Here is what PyAI covers and the practices schools should follow. Departments should also decide how long transcripts stay on the course page after the term ends.

Build a lecture transcription from your own code

Wire lecture transcription into your product through Hear in one of three ways. Full docs: PyAI SDKs.

SDK

Call Hear as a plain SDK function, send a recording and read back the transcript.

Audio and data come back ready to drop into the step that follows.

pip install pyai

MCP

Hand Hear to an assistant through the PyAI MCP server, so it runs your lecture transcription as a tool inside a chat.

You grant access once and can revoke it any time, and every call is logged.

pyai mcp add

CLI

Generate the recording notice audio for your lecture transcription from the terminal in a single line.

Each command returns audio in WAV, PCM or G.711, ready for a phone line.

pyai speak "..."

Lecture transcription FAQ

What is lecture transcription?

Lecture transcription is the process of turning the spoken audio of a class into written text. Schools use it for captions, searchable transcripts and study notes. With PyAI Hear, a recording or a live stream goes in and text with timestamps comes out, for $0.001/min of audio.

How do I turn a lecture to text?

Record the class with a clip on mic or the room system, then upload the file to PyAI Hear or stream it live through the API. Clear audio gives much better results, so keep the mic close to the speaker and cut background noise where you can.

Is class transcription accurate for course terms?

Accuracy depends most on audio quality. A close mic and a quiet room give clean text, while distant room mics drop words. Course specific terms and names are where general tools slip most, so review the first transcripts and correct terms before sharing them with students.

Do I need permission to record and transcribe a lecture?

Usually yes. Ask the lecturer before recording, tell students when a class is captured, and follow your school's recording policy and any disability services guidance. Trace can flag PII in transcripts before they are shared. Recordings and transcripts sit in encrypted storage scoped to your organization.

How much does lecture transcription cost?

PyAI Hear costs $0.001/min of audio. A department recording 120 hours of lectures a week uses about 28,800 minutes a month, which is about $29. New accounts get $50 in free credit with no credit card, and sandbox keys never bill.

Can PyAI caption a live lecture as it happens?

Yes. A live class can stream to PyAI Hear through the API, and text comes back with timestamps while the lecturer speaks. Your lecture capture system or caption player shows that text on screen, since PyAI does not provide a classroom player. Test the room mic first, because clear audio drives caption quality.

Start building with PyAI voice

Spin up a sandbox key and put a natural voice on your calls and apps in minutes. No credit card, and $50 in free credit.

Get a sandbox key →