omniai-openai 1.0.2 → 1.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 667d8318e96d3e6611a3ca20ced5dcefefc0896a4c84c48427ce441756cb9e60
-  data.tar.gz: 13161c5726cc33fcc7934a4164f751244b3c577fb52a1cdf30301dc10f733c84
+  metadata.gz: f35b8f18f1fed4fc62a4d69d1ec9d94b7b93dfe6d7110e9f64369625cafd8702
+  data.tar.gz: 746528fa3065f9f843bafb9b82946dda6d57bb121b81bbf6a8f4c43e72e1888e
 SHA512:
-  metadata.gz: d44485bdcebe5ee67a846c4517ea62fdb4fd98dfbab464f64fd10316561a396369ce5579367c8f073d79edfee0a790baa46c2fc31dfd2a4522221f0d16679a23
-  data.tar.gz: 2f1054f58e35f37012bc6b95f7605b0eeee01010718c2446a14b623b37cde0028d08ed56330d10bf1145fe5dfa69c3c138f54f5ec7d2bf65e55eb4727a21c664
+  metadata.gz: 4a11987f0c6dceeed121166820240c9c878ddc7f88b7c2cf2598c5f14ed5cdc1d6730d3463396f4c56b38f37d7ac4d983b7af461afd00c58db613a58b887d075
+  data.tar.gz: 129bd67690492743d3945fc15da22d5b9f4ec1bc384c56d8dcd2d60d1fd0419f802c6b804cbf154efdd4828d5e42220a30f890d58c08f9aaa634c1c8ba8bfa20

data/Gemfile

@@ -11,4 +11,5 @@ gem 'rspec_junit_formatter'
 gem 'rubocop'
 gem 'rubocop-rake'
 gem 'rubocop-rspec'
+gem 'simplecov'
 gem 'webmock'

data/README.md

@@ -116,3 +116,56 @@ JSON.parse(completion.choice.message.content) # { "name": "Ringo" }
 [OpenAI API Reference `response_format`](https://platform.openai.com/docs/api-reference/chat/create#chat-create-stream)
 
 > When using JSON mode, you must also instruct the model to produce JSON yourself via a system or user message.
+
+### Transcribe
+
+A transcription is generated by passing in a path to a file:
+
+```ruby
+transcription = client.transcribe(file.path)
+transcription.text # '...'
+```
+
+#### Prompt
+
+`prompt` is optional and can provide additional context for transcribing:
+
+```ruby
+transcription = client.transcribe(file.path, prompt: '')
+transcription.text # '...'
+```
+
+[OpenAI API Reference `prompt`](https://platform.openai.com/docs/api-reference/audio/createTranscription#audio-createtranscription-prompt)
+
+#### Format
+
+`format` is optional and supports `json`, `text`, `srt` or `vtt`:
+
+```ruby
+transcription = client.transcribe(file.path, format: OmniAI::Transcribe::Format::TEXT)
+transcription.text # '...'
+```
+
+[OpenAI API Reference `response_format`](https://platform.openai.com/docs/api-reference/audio/createTranscription#audio-createtranscription-response_format)
+
+#### Language
+
+`language` is optional and may improve accuracy and latency:
+
+```ruby
+transcription = client.transcribe(file.path, language: OmniAI::Transcribe::Language::SPANISH)
+transcription.text
+```
+
+[OpenAI API Reference `language`](https://platform.openai.com/docs/api-reference/audio/createTranscription#audio-createtranscription-language)
+
+#### Temperature
+
+`temperature` is optional and must be between 0.0 (more deterministic) and 1.0 (less deterministic):
+
+```ruby
+transcription = client.transcribe(file.path, temperature: 0.2)
+transcription.text
+```
+
+[OpenAI API Reference `temperature`](https://platform.openai.com/docs/api-reference/audio/createTranscription#audio-createtranscription-temperature)

data/lib/omniai/openai/client.rb

@@ -64,6 +64,20 @@ module OmniAI
       def chat(messages, model: Chat::Model::GPT_4O, temperature: nil, format: nil, stream: nil)
         Chat.process!(messages, model:, temperature:, format:, stream:, client: self)
       end
+
+      # @raise [OmniAI::Error]
+      #
+      # @param path [String]
+      # @param model [String]
+      # @param language [String, nil] optional
+      # @param prompt [String, nil] optional
+      # @param temperature [Float, nil] optional
+      # @param format [Symbol] :text, :srt, :vtt, or :json (default)
+      #
+      # @return text [OmniAI::Transcribe::Transcription]
+      def transcribe(path, model: Transcribe::Model::WHISPER, language: nil, prompt: nil, temperature: nil, format: nil)
+        Transcribe.process!(path, model:, language:, prompt:, temperature:, format:, client: self)
+      end
     end
   end
 end

data/lib/omniai/openai/config.rb

@@ -4,7 +4,7 @@ module OmniAI
   module OpenAI
     # Configuration for managing the OpenAI `api_key` / `organization` / `project` / `logger`.
     class Config < OmniAI::Config
-      attr_accessor :organization, :project, :chat_options
+      attr_accessor :organization, :project, :chat_options, :transcribe_options
 
       def initialize
         super
@@ -13,6 +13,7 @@ module OmniAI
         @project = ENV.fetch('OPENAI_PROJECT', nil)
         @host = ENV.fetch('OPENAI_HOST', 'https://api.openai.com')
         @chat_options = {}
+        @transcribe_options = {}
       end
     end
   end

data/lib/omniai/openai/transcribe.rb

@@ -0,0 +1,28 @@
+# frozen_string_literal: true
+
+module OmniAI
+  module OpenAI
+    # An OpenAI transcribe implementation.
+    class Transcribe < OmniAI::Transcribe
+      module Model
+        WHISPER_1 = 'whisper-1'
+        WHISPER = WHISPER_1
+      end
+
+      protected
+
+      # @return [Hash]
+      def payload
+        OmniAI::OpenAI
+          .config.transcribe_options
+          .merge(super)
+          .merge({ response_format: @format || Format::JSON })
+      end
+
+      # @return [String]
+      def path
+        "/#{OmniAI::OpenAI::Client::VERSION}/audio/transcriptions"
+      end
+    end
+  end
+end

data/lib/omniai/openai/version.rb

@@ -2,6 +2,6 @@
 
 module OmniAI
   module OpenAI
-    VERSION = '1.0.2'
+    VERSION = '1.1.0'
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: omniai-openai
 version: !ruby/object:Gem::Version
-  version: 1.0.2
+  version: 1.1.0
 platform: ruby
 authors:
 - Kevin Sylvestre
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2024-06-15 00:00:00.000000000 Z
+date: 2024-06-17 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: event_stream_parser
@@ -65,6 +65,7 @@ files:
 - lib/omniai/openai/chat.rb
 - lib/omniai/openai/client.rb
 - lib/omniai/openai/config.rb
+- lib/omniai/openai/transcribe.rb
 - lib/omniai/openai/version.rb
 homepage: https://github.com/ksylvest/omniai-openai
 licenses: []