> ## Documentation Index
> Fetch the complete documentation index at: https://assembly-preview.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Ask questions about your audio data

> In this guide, you'll learn how to use LeMUR to ask questions and get answers about your audio data.

<Tip>
  If you want a Quickstart, see [Apply LLMs to audio files](/docs/getting-started/apply-llms-to-audio-files).
</Tip>

<Note>
  To use LeMUR, you need an with a credit card set up.
</Note>

## Basic Q\&A example

If you want to send a custom prompt to the LLM, you can use the [LeMUR Task](https://assemblyai.com/docs/api-reference/lemur/task) and apply the model to your transcribed audio files.

To ask question about your audio data, define a prompt with your questions and call `transcript.lemur.task()`. The underlying `transcript` is automatically used as additional context for the model.

<CodeGroup>
  ```python python theme={"system"}
  import assemblyai as aai

  aai.settings.api_key = "YOUR_API_KEY"

  # Step 1: Transcribe an audio file.
  # audio_file = "./local_file.mp3"
  audio_file = "https://assembly.ai/sports_injuries.mp3"

  transcriber = aai.Transcriber()
  transcript = transcriber.transcribe(audio_file)

  # Step 2: Define a prompt with your question(s).
  prompt = "What is a runner's knee?"

  # Step 3: Apply LeMUR.
  result = transcript.lemur.task(
      prompt, final_model=aai.LemurModel.claude3_5_sonnet
  )

  print(result.response)
  ```

  ```typescript Typescript theme={"system"}
  import { AssemblyAI } from 'assemblyai'

  const client = new AssemblyAI({
    apiKey: 'YOUR_API_KEY'
  })

  const run = async () => {
    // Step 1: Transcribe an audio file.
    //const audioFile = './local_file.mp3'
    const audioFile =
      'https://assembly.ai/sports_injuries.mp3'
    const transcript = await client.transcripts.transcribe({ audio: audioFile })

    // Step 2: Define a prompt with your question(s).
    const prompt = "What is a runner's knee?"

    // Step 3: Apply LeMUR.
    const { response } = await client.lemur.task({
      transcript_ids: [transcript.id],
      prompt,
      final_model: 'anthropic/claude-3-5-sonnet'
    })

    console.log(response)
  }

  run()
  ```

  ```go Go theme={"system"}
  package main

  import (
      "context"
      "fmt"

      aai "github.com/AssemblyAI/assemblyai-go-sdk"
  )

  func main() {
      ctx := context.Background()

      client := aai.NewClient("YOUR_API_KEY")

      // Step 1: Transcribe an audio file. For local files see our Getting Started guides.
      audioURL := "https://assembly.ai/sports_injuries.mp3"
      transcript, _ := client.Transcripts.TranscribeFromURL(ctx, audioURL, nil)

      // Step 2: Define a prompt with your question.
      prompt := "What is a runner's knee?"

      // Step 3: Apply LeMUR.
      var params aai.LeMURTaskParams
      params.Prompt = aai.String(prompt)
      params.TranscriptIDs = []string{aai.ToString(transcript.ID)}
      params.FinalModel = "anthropic/claude-3-5-sonnet"

      result, _ := client.LeMUR.Task(ctx, params)

      fmt.Println(*result.Response)
  }
  ```

  ```java Java theme={"system"}
  import com.assemblyai.api.AssemblyAI;
  import com.assemblyai.api.resources.transcripts.types.*;
  import com.assemblyai.api.resources.lemur.requests.*;
  import java.util.List;

  public final class App {
      public static void main(String[] args) {

          AssemblyAI client = AssemblyAI.builder()
                  .apiKey("YOUR_API_KEY")
                  .build();

          // Step 1: Transcribe an audio file. For local files see our Getting Started guides.
          String audioUrl = "https://assembly.ai/sports_injuries.mp3";
          Transcript transcript = client.transcripts().transcribe(audioUrl);

          // Step 2: Define a prompt with your question(s).
          String prompt = "What is a runner's knee?";

          // Step 3: Apply LeMUR.
          var params = LemurTaskParams.builder()
                  .prompt(prompt)
                  .transcriptIds(List.of(transcript.getId()))
                  .finalModel(LemurModel.ANTHROPIC_CLAUDE3_5_SONNET)
                  .build();

          var response = client.lemur().task(params);

          System.out.println(response.getResponse());
      }
  }
  ```

  ```csharpt C# theme={"system"}
  using AssemblyAI;
  using AssemblyAI.Lemur;
  using AssemblyAI.Transcripts;

  var client = new AssemblyAIClient("YOUR_API_KEY");

  // Step 1: Transcribe an audio file. For local files see our Getting Started guides.
  var transcript = await client.Transcripts.TranscribeAsync(new TranscriptParams
  {
      AudioUrl = "https://assembly.ai/sports_injuries.mp3"
  });

  // Step 2: Define a prompt with your question(s).
  const string prompt = "What is a runner's knee?";

  // Step 3: Apply LeMUR.
  var lemurTaskParams = new LemurTaskParams
  {
      Prompt = prompt,
      TranscriptIds = [transcript.Id],
      FinalModel = LemurModel.AnthropicClaude3_5_Sonnet
  };

  var response = await client.Lemur.TaskAsync(lemurTaskParams);

  Console.WriteLine(response.Response);
  ```

  ```ruby Ruby theme={"system"}
  require 'assemblyai'

  client = AssemblyAI::Client.new(api_key: 'YOUR_API_KEY')

  # Step 1: Transcribe an audio file. For local files see our Getting Started guides.
  audio_url = 'https://assembly.ai/sports_injuries.mp3'
  transcript = client.transcripts.transcribe(audio_url: audio_url)

  # Step 2: Define a prompt with your question(s).
  prompt = "What is a runner's knee?"

  # Step 3: Apply LeMUR.
  response = client.lemur.task(
    prompt: prompt,
    transcript_ids: [transcript.id],
    final_model: AssemblyAI::Lemur::LemurModel::ANTHROPIC_CLAUDE3_5_SONNET
  )

  puts response.response
  ```
</CodeGroup>

#### Example output

```
Based on the transcript, runner's knee is a condition characterizedby pain behind or around the kneecap. It is caused by overuse,muscle imbalance and inadequate stretching. Symptoms include painunder or around the kneecap and pain when walking.
```

## Q\&A with specialized endpoint

The [LeMUR Question & Answer function](/docs/api-reference/lemur/question-answer) requires no prompt engineering and facilitates more deterministic and structured outputs. You can use it with `transcript.lemur.question()`.

To use it, define a list of `aai.LemurQuestion` objects. For each question, you can define additional `context` and specify either a `answer_format` or a list of `answer_options`. Additionally, you can define an overall `context`.

<CodeGroup>
  ```python python theme={"system"}
  import assemblyai as aai

  aai.settings.api_key = "YOUR_API_KEY"

  audio_url = "https://assembly.ai/meeting.mp4"
  transcript = aai.Transcriber().transcribe(audio_url)

  questions = [
      aai.LemurQuestion(
          question="What are the top level KPIs for engineering?",
          context="KPI stands for key performance indicator",
          answer_format="short sentence"),
      aai.LemurQuestion(
          question="How many days has it been since the data team has gotten updated metrics?",
          answer_options=["1", "2", "3", "4", "5", "6", "7", "more than 7"]),
  ]

  result = transcript.lemur.question(
      final_model=aai.LemurModel.claude3_5_sonnet,
      questions,
      context="A GitLab meeting to discuss logistics"
  )

  for qa_response in result.response:
      print(f"Question: {qa_response.question}")
      print(f"Answer: {qa_response.answer}")
  ```

  ```typescript Typescript theme={"system"}
  import { AssemblyAI } from 'assemblyai'

  const client = new AssemblyAI({
    apiKey: 'YOUR_API_KEY'
  })

  const audioUrl = 'https://assembly.ai/meeting.mp4'

  const run = async () => {
    const transcript = await client.transcripts.transcribe({ audio: audioUrl })

    const questions = [
      {
        question: 'What are the top level KPIs for engineering?',
        context: 'KPI stands for key performance indicator',
        answer_format: 'short sentence'
      },
      {
        question:
          'How many days has it been since the data team has gotten updated metrics?',
        answer_options: ['1', '2', '3', '4', '5', '6', '7', 'more than 7']
      }
    ]

    const { response: qas } = await client.lemur.questionAnswer({
      transcript_ids: [transcript.id],
      final_model: 'anthropic/claude-3-5-sonnet',
      context: 'A GitLab meeting to discuss logistics',
      questions: questions
    })

    for (const { question, answer } of qas) {
      console.log('Question', question)
      console.log('Answer', answer)
    }
  }

  run()
  ```

  ```java Java theme={"system"}
  import com.assemblyai.api.AssemblyAI;
  import com.assemblyai.api.resources.lemur.requests.*;
  import com.assemblyai.api.resources.lemur.types.*;
  import com.assemblyai.api.resources.transcripts.types.*;
  import java.util.List;

  public final class App {
      public static void main(String[] args) {

          AssemblyAI client = AssemblyAI.builder()
                  .apiKey("YOUR_API_KEY")
                  .build();

          String audioUrl = "https://assembly.ai/meeting.mp4";

          Transcript transcript = client.transcripts().transcribe(audioUrl);

          var question1 = LemurQuestion.builder()
                  .question("What are the top level KPIs for engineering?")
                  .context(LemurQuestionContext.of("KPI stands for key performance indicator"))
                  .answerFormat("short sentence").build();

          var question2 = LemurQuestion.builder()
                  .question("How many days has it been since the data team has gotten updated metrics?")
                  .answerOptions(List.of("1", "2", "3", "4", "5", "6", "7", "more than 7")).build();

          var response = client.lemur().questionAnswer(LemurQuestionAnswerParams.builder()
                  .transcriptIds(List.of(transcript.getId()))
                  .finalModel(LemurModel.ANTHROPIC_CLAUDE3_5_SONNET)
                  .context(LemurBaseParamsContext.of("A GitLab meeting to discuss logistic"))
                  .questions(List.of(question1, question2))
                  .build());

          for (var qa : response.getResponse()) {
              System.out.println("Question: " + qa.getQuestion());
              System.out.println("Answer: " + qa.getAnswer());
          }
      }
  }
  ```

  ```csharpt C# theme={"system"}
  using AssemblyAI;
  using AssemblyAI.Lemur;
  using AssemblyAI.Transcripts;

  var client = new AssemblyAIClient("YOUR_API_KEY");

  var transcript = await client.Transcripts.TranscribeAsync(new TranscriptParams
  {
      AudioUrl = "https://assembly.ai/meeting.mp4"
  });

  var lemurTaskParams = new LemurQuestionAnswerParams
  {
      TranscriptIds = [transcript.Id],
      FinalModel = LemurModel.AnthropicClaude3_5_Sonnet,
      Context = "A GitLab meeting to discuss logistic",
      Questions =
      [
          new LemurQuestion
          {
              Question = "What are the top level KPIs for engineering?",
              Context = "KPI stands for key performance indicator",
              AnswerFormat = "short sentence"
          },
          new LemurQuestion
          {
              Question = "How many days has it been since the data team has gotten updated metrics?",
              Context = "KPI stands for key performance indicator",
              AnswerOptions = ["1", "2", "3", "4", "5", "6", "7", "more than 7"]
          }
      ]
  };

  var response = await client.Lemur.QuestionAnswerAsync(lemurTaskParams);

  foreach (var qa in response.Response)
  {
      Console.WriteLine($"Question: {qa.Question}");
      Console.WriteLine($"Answer: {qa.Answer}");
  }
  ```

  ```ruby Ruby theme={"system"}
  require 'assemblyai'

  client = AssemblyAI::Client.new(api_key: 'YOUR_API_KEY')

  audio_url = 'https://assembly.ai/meeting.mp4'

  transcript = client.transcripts.transcribe(audio_url: audio_url)

  response = client.lemur.question_answer(
    transcript_ids: [transcript.id],
    final_model: AssemblyAI::Lemur::LemurModel::ANTHROPIC_CLAUDE3_5_SONNET,
    context: 'A GitLab meeting to discuss logistic',
    questions: [
      {
        question: 'What are the top level KPIs for engineering?',
        context: 'KPI stands for key performance indicator',
        answer_format: 'short sentence'
      },
      {
        question: 'How many days has it been since the data team has gotten updated metrics?',
        context: 'KPI stands for key performance indicator',
        answer_options: ['1', '2', '3', '4', '5', '6', '7', 'more than 7']
      }
    ]
  )

  response.response.each do |qa|
    printf("Question: %<question>s\n", question: qa.question)
    printf("Answer: %<answer>s\n", answer: qa.answer)
  end
  ```
</CodeGroup>

## Custom Q\&A example (Advanced)

This example shows how you can run a custom LeMUR task with an advanced prompt to create custom Q\&A responses:

[Cookbook: Custom Q\&A with LeMUR Task](https://github.com/AssemblyAI/cookbook/blob/master/lemur/task-endpoint-structured-QA.ipynb)

## More Q\&A prompt examples

Try any of these prompts to get started:

| Use case               | Example prompt                                                       |
| ---------------------- | -------------------------------------------------------------------- |
| Question and answer    | *"Identify any patterns or trends based on the transcript"*          |
| Closed-ended questions | *"Did the customer express a positive sentiment in the phone call?"* |
| Sentiment analysis     | *"What was the emotional sentiment of the phone call?"*              |

For more use cases and prompt examples, see [LeMUR examples](/docs/lemur/examples).

## API reference

* [LeMUR Task endpoint](https://assemblyai.com/docs/api-reference/lemur/task)
* [LeMUR Question & Answer endpoint](/docs/api-reference/lemur/question-answer)

## Improve the results

To improve the results, see the following resources:

* Optimize your prompt with the [prompt engineering guide](/docs/lemur/improving-your-prompt).
* To alter the outcome, see [Change model and parameters](/docs/lemur/customize-parameters).