omniai-openai 1.1.0 → 1.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f35b8f18f1fed4fc62a4d69d1ec9d94b7b93dfe6d7110e9f64369625cafd8702
-  data.tar.gz: 746528fa3065f9f843bafb9b82946dda6d57bb121b81bbf6a8f4c43e72e1888e
+  metadata.gz: 01d9c9144c74769c3eed7d8ec37f4cab63885057d5f16b02727e8c26e94414a0
+  data.tar.gz: 98e56e503c8ba01d3822d448dab53c94fef16d08c3f7896884e381f0b10b54fb
 SHA512:
-  metadata.gz: 4a11987f0c6dceeed121166820240c9c878ddc7f88b7c2cf2598c5f14ed5cdc1d6730d3463396f4c56b38f37d7ac4d983b7af461afd00c58db613a58b887d075
-  data.tar.gz: 129bd67690492743d3945fc15da22d5b9f4ec1bc384c56d8dcd2d60d1fd0419f802c6b804cbf154efdd4828d5e42220a30f890d58c08f9aaa634c1c8ba8bfa20
+  metadata.gz: 652b78a11ffb57f7b884d39fd08222fd4a87a0105a32ec0304d7d09efde5988b7f6058d3fbed46b6b3b8e8d44fec5191ffe925804f953456502020ea9cc8ef6f
+  data.tar.gz: 41c3516078e9a9411eebea6a0eaf3424b6a756d8b714697e51817e02f7faa9c4f170dc38c2fadb21c325a7576c5ae615468e3fb5d1438014532ae1be4c43d37b

data/README.md

@@ -1,5 +1,7 @@
 # OmniAI::OpenAI
 
+[![CircleCI](https://circleci.com/gh/ksylvest/omniai-openai.svg?style=svg)](https://circleci.com/gh/ksylvest/omniai-openai)
+
 An OpenAI implementation of the [OmniAI](https://github.com/ksylvest/omniai) APIs.
 
 ## Installation
@@ -43,12 +45,12 @@ end
 A chat completion is generated by passing in prompts using any a variety of formats:
 
 ```ruby
-completion = client.chat.completion('Tell me a joke!')
+completion = client.chat('Tell me a joke!')
 completion.choice.message.content # 'Why did the chicken cross the road? To get to the other side.'
 ```
 
 ```ruby
-completion = client.chat.completion({
+completion = client.chat({
   role: OmniAI::Chat::Role::USER,
   content: 'Is it wise to jump off a bridge?'
 })
@@ -56,7 +58,7 @@ completion.choice.message.content # 'No.'
 ```
 
 ```ruby
-completion = client.chat.completion([
+completion = client.chat([
   {
     role: OmniAI::Chat::Role::SYSTEM,
     content: 'You are a helpful assistant.'
@@ -71,7 +73,7 @@ completion.choice.message.content # 'The capital of Canada is Ottawa.'
 `model` takes an optional string (default is `gtp-4o`):
 
 ```ruby
-completion = client.chat.completion('How fast is a cheetah?', model: OmniAI::OpenAI::Chat::Model::GPT_3_5_TURBO)
+completion = client.chat('How fast is a cheetah?', model: OmniAI::OpenAI::Chat::Model::GPT_3_5_TURBO)
 completion.choice.message.content # 'A cheetah can reach speeds over 100 km/h.'
 ```
 
@@ -82,7 +84,7 @@ completion.choice.message.content # 'A cheetah can reach speeds over 100 km/h.'
 `temperature` takes an optional float between `0.0` and `2.0` (defaults is `0.7`):
 
 ```ruby
-completion = client.chat.completion('Pick a number between 1 and 5', temperature: 2.0)
+completion = client.chat('Pick a number between 1 and 5', temperature: 2.0)
 completion.choice.message.content # '3'
 ```
 
@@ -96,7 +98,7 @@ completion.choice.message.content # '3'
 stream = proc do |chunk|
   print(chunk.choice.delta.content) # 'Better', 'three', 'hours', ...
 end
-client.chat.completion('Be poetic.', stream:)
+client.chat('Be poetic.', stream:)
 ```
 
 [OpenAI API Reference `stream`](https://platform.openai.com/docs/api-reference/chat/create#chat-create-stream)
@@ -106,7 +108,7 @@ client.chat.completion('Be poetic.', stream:)
 `format` takes an optional symbol (`:json`) and that setes the `response_format` to `json_object`:
 
 ```ruby
-completion = client.chat.completion([
+completion = client.chat([
   { role: OmniAI::Chat::Role::SYSTEM, content: OmniAI::Chat::JSON_PROMPT },
   { role: OmniAI::Chat::Role::USER, content: 'What is the name of the drummer for the Beatles?' }
 ], format: :json)
@@ -169,3 +171,63 @@ transcription.text
 ```
 
 [OpenAI API Reference `temperature`](https://platform.openai.com/docs/api-reference/audio/createTranscription#audio-createtranscription-temperature)
+
+### Speak
+
+Speech can be generated by passing text with a block:
+
+```ruby
+File.open('example.ogg', 'wb') do |file|
+  client.speak('How can a clam cram in a clean cream can?') do |chunk|
+    file << chunk
+  end
+end
+```
+
+If a block is not provided then a tempfile is returned:
+
+```ruby
+tempfile = client.speak('Can you can a can as a canner can can a can?')
+tempfile.close
+tempfile.unlink
+```
+
+#### Voice
+
+`voice` is optional and must be one of the supported voices:
+
+```ruby
+client.speak('She sells seashells by the seashore.', voice: OmniAI::OpenAI::Speak::Voice::SHIMMER)
+```
+
+[OpenAI API Reference `voice`](https://platform.openai.com/docs/api-reference/audio/createSpeech#audio-createspeech-voice)
+
+#### Model
+
+`model` is optional and must be either `tts-1` or `tts-1-hd` (default):
+
+```ruby
+client.speak('I saw a kitten eating chicken in the kitchen.', format: OmniAI::OpenAI::Speak::Model::TTS_1)
+```
+
+[OpenAI API Refernce `model`](https://platform.openai.com/docs/api-reference/audio/createSpeech#audio-createspeech-model)
+
+#### Speed
+
+`speed` is optional and must be between 0.25 and 0.40:
+
+```ruby
+client.speak('How much wood would a woodchuck chuck if a woodchuck could chuck wood?', speed: 4.0)
+```
+
+[OmniAI API Reference `speed`](https://platform.openai.com/docs/api-reference/audio/createSpeech#audio-createspeech-speed)
+
+#### Format
+
+`format` is optional and supports `MP3` (default), `OPUS`, `AAC`, `FLAC`, `WAV` or `PCM`:
+
+```ruby
+client.speak('A pessemistic pest exists amidst us.', format: OmniAI::OpenAI::Speak::Format::FLAC)
+```
+
+[OpenAI API Reference `format`](https://platform.openai.com/docs/api-reference/audio/createSpeech#audio-createspeech-response_format)

data/lib/omniai/openai/client.rb

@@ -78,6 +78,27 @@ module OmniAI
       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
+
+      # @raise [OmniAI::Error]
+      #
+      # @param input [String] required
+      # @param model [String] optional
+      # @param voice [String] optional
+      # @param speed [Float] optional
+      # @param format [String] optional (default "aac"):
+      #   - "aac"
+      #   - "mp3"
+      #   - "flac"
+      #   - "opus"
+      #   - "pcm"
+      #   - "wav"
+      #
+      # @yield [output] optional
+      #
+      # @return [Tempfile``]
+      def speak(input, model: Speak::Model::TTS_1_HD, voice: Speak::Voice::ALLOY, speed: nil, format: nil, &)
+        Speak.process!(input, model:, voice:, speed:, 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, :transcribe_options
+      attr_accessor :organization, :project, :chat_options, :transcribe_options, :speak_options
 
       def initialize
         super
@@ -14,6 +14,7 @@ module OmniAI
         @host = ENV.fetch('OPENAI_HOST', 'https://api.openai.com')
         @chat_options = {}
         @transcribe_options = {}
+        @speak_options = {}
       end
     end
   end

data/lib/omniai/openai/speak.rb

@@ -0,0 +1,37 @@
+# frozen_string_literal: true
+
+module OmniAI
+  module OpenAI
+    # An OpenAI transcribe implementation.
+    class Speak < OmniAI::Speak
+      module Model
+        TTS_1 = 'tts-1'
+        TTS_1_HD = 'tts-1-hd'
+      end
+
+      module Voice
+        ALLOY = 'alloy' # https://platform.openai.com/docs/guides/text-to-speech/alloy
+        ECHO = 'echo' # https://platform.openai.com/docs/guides/text-to-speech/echo
+        FABLE = 'fable' # https://platform.openai.com/docs/guides/text-to-speech/fable
+        NOVA = 'nova' # https://platform.openai.com/docs/guides/text-to-speech/nova
+        ONYX = 'onyx' # https://platform.openai.com/docs/guides/text-to-speech/onyx
+        SHIMMER = 'shimmer' # https://platform.openai.com/docs/guides/text-to-speech/shimmer
+      end
+
+      protected
+
+      # @return [Hash]
+      def payload
+        OmniAI::OpenAI
+          .config.speak_options
+          .merge(super)
+          .merge({ response_format: @format }.compact)
+      end
+
+      # @return [String]
+      def path
+        "/#{OmniAI::OpenAI::Client::VERSION}/audio/speech"
+      end
+    end
+  end
+end

data/lib/omniai/openai/version.rb

@@ -2,6 +2,6 @@
 
 module OmniAI
   module OpenAI
-    VERSION = '1.1.0'
+    VERSION = '1.2.0'
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: omniai-openai
 version: !ruby/object:Gem::Version
-  version: 1.1.0
+  version: 1.2.0
 platform: ruby
 authors:
 - Kevin Sylvestre
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2024-06-17 00:00:00.000000000 Z
+date: 2024-06-20 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: event_stream_parser
@@ -65,10 +65,12 @@ files:
 - lib/omniai/openai/chat.rb
 - lib/omniai/openai/client.rb
 - lib/omniai/openai/config.rb
+- lib/omniai/openai/speak.rb
 - lib/omniai/openai/transcribe.rb
 - lib/omniai/openai/version.rb
 homepage: https://github.com/ksylvest/omniai-openai
-licenses: []
+licenses:
+- MIT
 metadata:
   homepage_uri: https://github.com/ksylvest/omniai-openai
   changelog_uri: https://github.com/ksylvest/omniai-openai/releases