Universal-3 Pro (Async)

Universal-3 Pro is our most powerful Voice AI model, designed to capture the “hard stuff” that traditional ASR models struggle with. It delivers state-of-the-art accuracy for entities, rare words, and domain-specific terminology out of the box, with prompting support for fully customized transcription output.

Quickstart

Get started with Universal-3 Pro using the code below. This example transcribes a pre-recorded audio file and prints the transcript text to your terminal.

$
pip install "assemblyai>=1.0.0"

1
import assemblyai as aai
2
3
aai.settings.api_key = "<YOUR_API_KEY>"
4
5
audio_file = "https://assembly.ai/sports_injuries.mp3"
6
7
config = aai.TranscriptionConfig(
8
  speech_models=["universal-3-pro", "universal-2"],
9
  language_detection=True,
10
)
11
12
transcript = aai.Transcriber().transcribe(audio_file, config)
13
14
print(transcript.text)

Key capabilities

The model out of the box outperforms all ASR models on the market on accuracy, especially as it pertains to entities and rare words. With prompting, you can get an entirely customized transcription output that rivals near-human-level transcription.

• Keyterm Prompting: Improve recognition of domain-specific terminology, rare words, and proper nouns
• Prompting: Guide transcription style, formatting, and output characteristics

To fine-tune to your use case, see the Prompting section. Not sure where to start? Try our Prompt Generator.

Start with no prompt

We strongly recommend testing with no prompt first. When you omit the prompt parameter, Universal-3-Pro automatically applies a built-in default prompt that is already optimized for accuracy across a wide range of audio types — including verbatim transcription, multilingual code-switching, and challenging audio conditions. For most use cases, the default prompt delivers excellent results out of the box.

If you’re going to build a prompt, start with one of the recommended prompts and then tweak it for your use case. You should not start from scratch with your prompt — use a recommended prompt and then build off of it. Please read the Prompting Guide (Async) if you’d like to build your prompt yourself.

Remember, prompts are primarily instructional, so adding a large amount of context may not make a significant impact on accuracy and could reduce instruction-following coherence. Feel free to layer in additional instructions that you see in the Prompting Guide (Async).

Benchmarking

For more details on evaluating transcription accuracy, including tips on using semantic WER and handling substitution artifacts, see the Evaluating transcription accuracy section in the Prompting Guide.

Keyterms prompting

Keyterms prompting allows you to provide up to 1,000 words or phrases (maximum 6 words per phrase) using the keyterms_prompt parameter to improve transcription accuracy for those terms and related variations or contextually similar phrases.

Here is an example showing how you can use keyterms prompting to improve transcription accuracy for a name with distinctive spelling and formatting.

Without keyterms prompting:

Hi, this is Kelly Byrne Donahue

With keyterms prompting:

Hi, this is Kelly Byrne-Donoghue

1
import assemblyai as aai
2
3
aai.settings.api_key = "<YOUR_API_KEY>"
4
5
audio_file = "https://assemblyaiassets.com/audios/keyterms_prompting.wav"
6
7
config = aai.TranscriptionConfig(
8
  speech_models=["universal-3-pro", "universal-2"],
9
  language_detection=True,
10
  keyterms_prompt=["Kelly Byrne-Donoghue"],
11
)
12
13
transcript = aai.Transcriber().transcribe(audio_file, config)
14
15
print(transcript.text)

How keyterms prompting works behind the scenes

Behind the scenes, the keyterms_prompt parameter works by appending your keyterms to the system prompt. The model receives the combined <prompt> + <keyterms> as a single input, which is how keyterms are boosted during transcription.

You are free to replicate this behavior manually by appending keyterms directly to the end of your prompt parameter. For example:

Always: Transcribe speech with your best guess based on context in all possible scenarios where speech is present in the audio. Make sure to boost the words Kelly Byrne-Donoghue, AssemblyAI, cardiology in the audio.

Prompting

For a comprehensive guide on crafting effective prompts, including best practices, prompt capabilities, and example prompts, see the Prompting guide.

Universal-3 Pro delivers great accuracy out of the box. To fine-tune transcription results to your use case, provide a prompt with up to 1,500 words of context in plain language. This helps the model consistently recognize domain-specific terminology, apply your preferred formatting conventions, handle code switching between languages, and better interpret ambiguous speech.

Default prompt

When no prompt is provided, Universal-3 Pro automatically applies a default prompt based on the audio length:

• If the audio is longer than 10 seconds, the following default prompt is used:

Required: Preserve the original language(s) and script as spoken, including code-switching and mixed-language phrases.
Mandatory: Preserve linguistic speech patterns including disfluencies, filler words, hesitations, repetitions, stutters, false starts, and colloquialisms in the spoken language.
Always: Transcribe speech with your best guess based on context in all possible scenarios where speech is present in the audio.

• Otherwise, the default prompt is Transcribe verbatim.

You can override the default prompt by providing your own prompt value. See the sections below for examples.

Best simple prompt

If you want strong accuracy without the verbatim and multilingual instructions, use this single-line prompt:

Always: Transcribe speech with your best guess based on context in all possible scenarios where speech is present in the audio.

This gives the model clear guidance to always attempt transcription while keeping instructions minimal. It’s a great starting point if you don’t need multilingual code-switching or maximum disfluency capture.

Handling unclear audio with [masked]

Always: Transcribe speech exactly as heard. If uncertain or audio is unclear, mark as [masked]. After the first output, review the transcript again. Pay close attention to hallucinations, misspellings, or errors, and revise them like a computer performing spell and grammar checks. Ensure words and phrases make grammatical sense in sentences.

You can also use [unclear] instead of [masked].

The [masked] strategy materially reduces hallucinations while still preserving difficult but genuine speech. It also provides greater transparency in the transcript — sections where audio is unclear are explicitly flagged as [masked], giving you visibility into areas of uncertainty rather than forcing potentially incorrect guesses.

For more recommended prompts and detailed prompt capabilities, see the Prompting guide.

Verbatim transcription and disfluencies

Capture natural speech patterns exactly as spoken, including um, uh, false starts, repetitions, stutters. Add examples of the verbatim elements you want to transcribe in the prompt parameter to guide the model.

Without prompt:

Do you and Quentin still socialize when you come to Los Angeles, or is it like he's so used to having you here? No, no, no, we're friends. What do you do with him?

With prompt, the model better captures filler words like “uh” and false starts like “we, we, we’re friends”.

Do you and Quentin still socialize, uh, when you come to Los Angeles, or is it like he's so used to having you here? No, no, no, we, we, we're friends. What do you do with him?

1
import assemblyai as aai
2
3
aai.settings.api_key = "<YOUR_API_KEY>"
4
5
audio_file = "https://assemblyaiassets.com/audios/verbatim.mp3"
6
7
config = aai.TranscriptionConfig(
8
  speech_models=["universal-3-pro", "universal-2"],
9
  language_detection=True,
10
  prompt="Produce a transcript suitable for conversational analysis. Every disfluency is meaningful data. Include: fillers (um, uh, er, ah, hmm, mhm, like, you know, I mean), repetitions (I I, the the), restarts (I was- I went), stutters (th-that, b-but, no-not), and informal speech (gonna, wanna, gotta)",
11
)
12
13
transcript = aai.Transcriber().transcribe(audio_file, config)
14
15
print(transcript.text)

Example prompts:

Include spoken filler words like "um," "uh," "you know," "like," plus repetitions
and false starts when clearly spoken.

Preserve all disfluencies exactly as spoken including verbal hesitations,
restarts, and self-corrections (um, uh, I—I mean).

Transcribe verbatim:
- Fillers: yes (um, uh, like, you know)
- Repetitions: yes (I I, the the the)
- Stutters: yes (th-that, b-but)
- False starts: yes (I was— I went)
- Colloquial: yes (gonna, wanna, gotta)

Audio event tags

Audio tags capture non-speech events like music, laughter, pauses, applause, background noise, and other sounds in your audio. Include examples of audio tags you want to transcribe in the prompt parameter to guide the model.

Without prompt:

Your call has been forwarded to an automatic voice message system. At the tone, please record your message. When you have finished recording, you may hang up or press 1 for more options.

With prompting, non-speech events like beeps are called out in the transcript.

Your call has been forwarded to an automatic voice message system. At the tone, please record your message. When you have finished recording, you may hang up or press 1 for more options. [beep]

Here are some examples of audio tags you can prompt for: [music], [laugher], [applause], [noise], [pause], [inaudible], [sigh], [gasp], [cheering], [sound], [screaming], [bell], [beep], [sound effect], [buzzer], and more.

1
import assemblyai as aai
2
3
aai.settings.api_key = "<YOUR_API_KEY>"
4
5
audio_file = "https://assemblyaiassets.com/audios/audio_tag.mp3"
6
7
config = aai.TranscriptionConfig(
8
  speech_models=["universal-3-pro", "universal-2"],
9
  language_detection=True,
10
  prompt="Produce a transcript suitable for conversational analysis. Every disfluency is meaningful data. Include: Tag sounds: [beep]",
11
)
12
13
transcript = aai.Transcriber().transcribe(audio_file, config)
14
15
print(transcript.text)

Labeling crosstalk

Label overlapping speech, interruptions, and crosstalk segments in the transcript.

Without prompt:

I hope you got our card. Okay, nobody talk. We'll just wait for her to talk. Well, we just wanted to— damn it!

With prompt:

I hope you got our card. [CROSSTALK] Okay, nobody talk. We'll just wait for her to talk. Well, we just wanted to— [CROSSTALK] Damn it!

1
import assemblyai as aai
2
3
aai.settings.api_key = "<YOUR_API_KEY>"
4
5
audio_file = "https://assemblyaiassets.com/audios/Difficult_audio.mp3"
6
7
config = aai.TranscriptionConfig(
8
  speech_models=["universal-3-pro", "universal-2"],
9
  language_detection=True,
10
  prompt="When multiple speakers talk simultaneously, mark crosstalk segments.",
11
)
12
13
transcript = aai.Transcriber().transcribe(audio_file, config)
14
15
print(transcript.text)

Example prompts:

If unintelligible, write (unclear).

Mark inaudible segments. Preserve overlapping speech and crosstalk.

Include pause markers where the speaker hesitates significantly.

Speaker attribution

Prompted speaker diarization identifies who said what in a conversation. It works especially well in cases where there are frequent interjections, such as quick acknowledgments or single-word responses, or when working with limited spoken audio, such as short-duration files.

Without prompt:

Speaker A: Five milligrams. And you take it regularly? Good. Every evening. And no side effects with it.

With prompt:

[Speaker:NURSE] 5mg. And do you take it regularly?
[Speaker:PATIENT] Oh yeah, yeah.
[Speaker:NURSE] Good.
[Speaker:PATIENT] I take it every evening.
[Speaker:NURSE] And no side effects with it?

Without prompting, it may appear that speaker B said everything. But with prompting, the model correctly identifies this as 5 separate speaker turns, capturing utterances as short as a single word, like “good”.

1
import assemblyai as aai
2
3
aai.settings.api_key = "<YOUR_API_KEY>"
4
5
audio_file = "https://assemblyaiassets.com/audios/speaker_diarization.mp3"
6
7
config = aai.TranscriptionConfig(
8
  speech_models=["universal-3-pro", "universal-2"],
9
  language_detection=True,
10
  speaker_labels=True,
11
  prompt="Produce a transcript with every disfluency data. Additionally, label speakers with their respective roles. 1. Place [Speaker:role] at the start of each speaker turn. Example format: [Speaker:NURSE] Hello there. How can I help you today? [Speaker:PATIENT] I'm feeling unwell. I have a headache.",
12
)
13
14
transcript = aai.Transcriber().transcribe(audio_file, config)
15
16
for utterance in transcript.utterances:
17
  print(f"Speaker {utterance.speaker}: {utterance.text}")

Output style and formatting

Control punctuation, capitalization, and readability without changing words.

Without prompt:

You got called because you were being loud and screaming. No, that's literally what my dispatch said. I don't give a fuck what your dispatch said. They lied. Okay, well, you need to calm down. I don't. Okay, yeah, calm down please. No, I don't. Yes, I'm Jesus Christ's daughter. I'm not doing this tonight with you. I'm not. I'm not. So you need to calm down.

With prompt, the model accurately captures the speaker’s emotional state through punctuation, adding exclamation marks during moments of yelling and emphasis.

You got called because you were being loud and screaming. No, I wasn't. That's literally what my dispatch said. I don't give a fuck what your dispatch said! They lied! Okay, well, you need to calm down. I don't! Okay, yeah, calm down, please. No, I don't! I'm Jesus Christ's daughter! I'm not doing this tonight with you. I'm not. I'm not. So you need to calm down.

1
import assemblyai as aai
2
3
aai.settings.api_key = "<YOUR_API_KEY>"
4
5
audio_file = "https://assemblyaiassets.com/audios/ouput_formatting.mp3"
6
7
config = aai.TranscriptionConfig(
8
  speech_models=["universal-3-pro", "universal-2"],
9
  language_detection=True,
10
  prompt="Add punctuation based on the speaker's tone and expressiveness. Use exclamation marks (!) when the speaker is yelling, excited, or emphatic. Use question marks (?) for questioning intonation. Apply standard punctuation (periods, commas) based on natural speech patterns and pauses.",
11
)
12
13
transcript = aai.Transcriber().transcribe(audio_file, config)
14
15
print(transcript.text)

Example prompts:

Transcribe this audio with beautiful punctuation and formatting.

Use expressive punctuation to reflect emotion and prosody.

Use standard punctuation and sentence breaks for readability.

Numbers and measurements

Control how numbers, percentages, and measurements are formatted.

Without prompt:

Commission has presented their communication, a hydrogen strategy for climate-neutral Europe, two weeks ago, which includes investments of between €180 billion and €400 billion.

With prompt:

Commission has presented their communication, a hydrogen strategy for climate-neutral Europe, 2 weeks ago, which includes investments of between €180,000,000,000 and €400,000,000,000.

1
import assemblyai as aai
2
3
aai.settings.api_key = "<YOUR_API_KEY>"
4
5
audio_file = "https://assemblyaiassets.com/audios/numbers_formatting.mp3"
6
7
config = aai.TranscriptionConfig(
8
  speech_models=["universal-3-pro", "universal-2"],
9
  language_detection=True,
10
  prompt="Transcribe with numbers normalized to standard formats. For example, when you see $1 billion, convert to $1,000,000,000.",
11
)
12
13
transcript = aai.Transcriber().transcribe(audio_file, config)
14
15
print(transcript.text)

Example prompts:

Convert spoken numbers to digits.

Use digits for numbers, percentages, and measurements.

Format financial figures with standard notation (Q3 revenue over third quarter revenue).

Context aware clues

Help with jargon, names, and domain expectations that are known from the audio file.

Without prompt:

I just want to move you along a bit further. Do you take any prescribed medicines? I know you've got diabetes and high blood pressure. I do. I take Ramipril. Okay. And I take Metformin, and there's another one that begins with G for the diabetes. Glicoside. Excellent. Okay. And do you know the dosage of the Ramipril? I think it's 5mg. 5mg, and do you take it regularly? Oh yeah, yeah. Good. Every evening. And no side effects with it? I did have an awful cough to start with, but that's stopped now. It's settled, good. Okay, and do you know the metformin, do you take it, do you know the dosage and how often? I take it 3 times a day, but I don't know what dose that is. Okay, we can check that out, we can follow that up, okay.

With prompt, adding ‘clinical history evaluation’ as a context clue corrects spelling of ‘Glicoside’ to ‘Glycoside’.

I just wanna move you along a bit further. Do you take any prescribed medicines? I know you've got diabetes and high blood pressure. I, I do. I take, um, I take Ramipril. Okay, mhm. And I take Metformin, and there's another one that begins with G for the diabetes. So glycosi— glycosi— glycoside. Excellent. Okay, and do you know the dosage of the Ramipril? Uh, I think it's 5mg. 5mg. And do you take it regularly? Oh yeah, yeah. Good. Every evening. And no side effects with it? Um, I did have an awful cough to start with, but that's, that's stopped now. It's settled, good. Okay, and do you know, erm, the metformin? Do you take it, do you know the dosage and how often? I take it 3 times a day, but I don't know what dose that is. Okay, we can check that out, we can follow that up, okay.

1
import assemblyai as aai
2
3
aai.settings.api_key = "<YOUR_API_KEY>"
4
5
audio_file = "https://assemblyaiassets.com/audios/nlp_prompting.mp3"
6
7
config = aai.TranscriptionConfig(
8
  speech_models=["universal-3-pro", "universal-2"],
9
  language_detection=True,
10
  prompt="Produce a transcript for a clinical history evaluation. It's important to capture medication and dosage accurately. Every disfluency is meaningful data. Include: fillers (um, uh, er, erm, ah, hmm, mhm, like, you know, I mean), repetitions (I I I, the the), restarts (I was- I went), stutters (th-that, b-but, no-not), and informal speech (gonna, wanna, gotta)",
11
)
12
13
transcript = aai.Transcriber().transcribe(audio_file, config)
14
15
print(transcript.text)

Example prompts:

Transcribe this audio. Context: a customer testimonial about contact center software.

Transcribe this audio. Context: a medical consultation discussing medications and symptoms.

Transcribe this audio. Context: a technical lecture about GPUs, CUDA, and inference.

Entity accuracy and spelling

Improve transcript accuracy for proper nouns, brands, technical terms, and domain vocabulary with prompting.

Without prompt:

Watch again closely. This is the potential game changer. The first responder NK cell killing cancer right before your eyes. If you give yourself Entiva, even in healthy volunteers, it dries up your first responders. It dries up your protectors. And that's why I said the power is within us.

With prompt, the model corrects the misrecognition of “Anktiva,” which would otherwise be transcribed as “Entiva”.

Watch again closely. This is the potential game changer. The first responder NK cell killing cancer right before your eyes. If you give yourself Anktiva, even in healthy volunteers, it dries up your first responders. It dries up your protectors. And that's why I said the power is within us.

1
import assemblyai as aai
2
3
aai.settings.api_key = "<YOUR_API_KEY>"
4
5
audio_file = "https://assemblyaiassets.com/audios/entity_accuracy.mp3"
6
7
config = aai.TranscriptionConfig(
8
  speech_models=["universal-3-pro", "universal-2"],
9
  language_detection=True,
10
  prompt="The speaker is discussing the cancer drug Anktiva (spelled A-N-K-T-I-V-A). When you hear what sounds like Entiva or similar pronunciations, transcribe it as Anktiva. This is the correct pharmaceutical name.",
11
)
12
13
transcript = aai.Transcriber().transcribe(audio_file, config)
14
15
print(transcript.text)

Example prompts:

Use standard spelling and the most contextually correct spelling of all words
including names, brands, drug names, medical terms, and proper nouns.

Non-negotiable: Pharmaceutical accuracy required (omeprazole over omeprizole,
metformin over metforman).
Preserve acronyms and capitalization (EBITDA over ebitda, API over A.P.I.).

Native code switching

Handle audio where speakers switch between languages.

Without prompt:

You literally lost your French? No, no, no. My French is there. Italian I've forgotten, but my French is still there and it will never leave. Okay, but would you need a French coach? Could you consider having me? Oh yeah, yeah, absolutely, absolutely. But for now, my French is there, fortunately.

With prompt, the model is able to preserve the speaker’s natural code switching between English and French, transcribing each language as spoken.

You literally lost your French? No, no, no. Mon français est là. L'italien, j'ai oublié, mais mon français est toujours là. Il partira jamais. Okay, but would you need a French coach? Could you consider having me? Oh yeah, yeah, absolutely, absolutely. Mais pour l'instant, le français est là, heureusement.

1
import assemblyai as aai
2
3
aai.settings.api_key = "<YOUR_API_KEY>"
4
5
audio_file = "https://assemblyaiassets.com/audios/code_switching_multilingual.mp3"
6
7
config = aai.TranscriptionConfig(
8
  speech_models=["universal-3-pro", "universal-2"],
9
  language_detection=True,
10
  prompt="The spoken language may change throughout the audio, transcribe in the original language mix (code-switching), preserving the words in the language they are spoken.",
11
)
12
13
transcript = aai.Transcriber().transcribe(audio_file, config)
14
15
print(transcript.text)

Example prompts:

Transcribe in the original language mix (code-switching), preserving words in the language they are spoken.

Preserve natural code-switching between English and Spanish. Retain spoken language as-is (correct "I was hablando con mi manager").

Temperature parameter

Control the amount of randomness injected into the model’s response using the temperature parameter.

The temperature parameter accepts values from 0.0 to 1.0, with a default value of 0.0.

1
import assemblyai as aai
2
3
aai.settings.api_key = "<YOUR_API_KEY>"
4
5
audio_file = "https://assemblyaiassets.com/audios/nlp_prompting.mp3"
6
7
config = aai.TranscriptionConfig(
8
  speech_models=["universal-3-pro", "universal-2"],
9
  language_detection=True,
10
  prompt="Produce a transcript for a clinical history evaluation. It's important to capture medication and dosage accurately. Every disfluency is meaningful data. Include: fillers (um, uh, er, erm, ah, hmm, mhm, like, you know, I mean), repetitions (I I I, the the), restarts (I was- I went), stutters (th-that, b-but, no-not), and informal speech (gonna, wanna, gotta)",
11
  temperature=0.1,
12
)
13
14
transcript = aai.Transcriber().transcribe(audio_file, config)
15
16
print(transcript.text)

Best practices for prompt engineering

Check out this guide to learn even more about how to craft effective prompts for Universal-3 Pro speech transcription, which includes an AI prompt generator tool.

Support for 99 languages

With the speech_models parameter, you can list multiple speech models in priority order, allowing our system to automatically route your audio based on language support.

Model routing behavior: The system attempts to use the models in priority order falling back to the next model when needed. For example, with ["universal-3-pro", "universal-2"], the system will try to use universal-3-pro for languages it supports (English, Spanish, Portuguese, French, German, and Italian), and automatically fall back to Universal-2 for all other languages. This ensures you get the best performing transcription where available while maintaining the widest language coverage.