cloudflare-ai 0.7.0 → 0.8.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 2f16a1fd5c112fb69440d4d89b1f27d167e854c537ca952b91e1f3f673c0efb5
-  data.tar.gz: 59f999e994f5bd21071ee2f125b5e138ba353feb857912b1df472f35acc05006
+  metadata.gz: dfe55c49c1310aa3940e9bc8963535680a93ccf90a0b98061b26a75c85fb42d3
+  data.tar.gz: 79335c8d797ca3585f8fb7a770fd064ac3aa7e76622fd8b794fbbb88abf24b43
 SHA512:
-  metadata.gz: 1d258b49887ca664f616aad4dbc74afc4bb8d49d0452933c2ee907306af56cf81da0bf084c260ddbe4db9d96f522812cc5abfb771c3d94757f12342334b16fd6
-  data.tar.gz: 133fcfd21cedbdb324604ec0e9bd4aacb0750b9f486df7bc217998e4dd117f4e6d482572ba1feee8b629e3362857cb8b8f28d2f1e3c6f1d89acddd18b73752e9
+  metadata.gz: ee82ccc27724c218043e4702a1dd39ba0878c6ff35caa35592e9dae3b26ea7bd1ccb0b73b182ad97864d637414edcfb1a587ec4b203fc4fe30767f15d56f4d74
+  data.tar.gz: 4cab183f89a4fd13cca87e8b0e264c93eddae4bf12c62a7ccfb5ae9e453b4cabc6916a3fdbf573ca55291ee8c537df190ebae06dbb2e2377af208e10cca8fe0c

data/README.md

@@ -14,18 +14,14 @@ generation to make legal services more accessible. [Email me](mailto:cloudflare-
 
 If you're looking for legal help, it's best to book a slot via https://www.krishnan.ca.
 
-# Todo
-It's still early days, and here are my immediate priorities:
-* [x] Support for streamed responses
-* [x] CI pipeline
-* [ ] Support for more AI model categories
-  * [x] [Text Generation](https://developers.cloudflare.com/workers-ai/models/text-generation/)
-  * [x] [Text Embeddings](https://developers.cloudflare.com/workers-ai/models/text-embeddings/)
-  * [x] [Text Classification](https://developers.cloudflare.com/workers-ai/models/text-classification/)
-  * [x] [Translation](https://developers.cloudflare.com/workers-ai/models/translation/)
-  * [x] [Image Classification](https://developers.cloudflare.com/workers-ai/models/image-classification/)
-  * [x] [Text-to-Image](https://developers.cloudflare.com/workers-ai/models/text-to-image/)
-  * [ ] [Automatic Speech Recognition](https://developers.cloudflare.com/workers-ai/models/speech-recognition/)
+# Supported features
+* [x] [Text Generation](https://developers.cloudflare.com/workers-ai/models/text-generation/)
+* [x] [Text Embeddings](https://developers.cloudflare.com/workers-ai/models/text-embeddings/)
+* [x] [Text Classification](https://developers.cloudflare.com/workers-ai/models/text-classification/)
+* [x] [Translation](https://developers.cloudflare.com/workers-ai/models/translation/)
+* [x] [Image Classification](https://developers.cloudflare.com/workers-ai/models/image-classification/)
+* [x] [Text-to-Image](https://developers.cloudflare.com/workers-ai/models/text-to-image/)
+* [x] [Automatic Speech Recognition](https://developers.cloudflare.com/workers-ai/models/speech-recognition/)
 
 # Table of Contents
 
@@ -180,9 +176,23 @@ All invocations of the `draw` method returns a `Cloudflare::AI::Results::TextToI
 result = client.translate(text: "Hello Jello", source_lang: "en", target_lang: "fr")
 p result.translated_text # => Hola Jello
 ```
+#### Result object
+All invocations of the `translate` method returns a `Cloudflare::AI::Results::Translate`.
+
 
+### Automatic speech recognition
+You can pass either a URL (source_url:) or a file (audio:) to the `transcribe` method. 
+```ruby
+result = client.transcribe(source_url: "http://example.org/path/to/audio.wav")
+p result.text # => "Hello Jello."
+p result.word_count # => 2
+p result.to_json # => {"result":{"text":"Hello Jello.","word_count":2,"words":[{"word":"Hello","start":0,"end":1.340000033378601},{"word":"Jello.","start":1.340000033378601,"end":1.340000033378601}},"success":true,"errors":[],"messages":[]}
+
+result = client.transcribe(audio: File.open("/path/to/audio.wav"))
+# ...
+```
 #### Result object
-All invocations of the `translate` methods return a `Cloudflare::AI::Results::Translate`.
+All invocations of the `transcribe` method returns a `Cloudflare::AI::Results::Transcribe`.
 
 # Logging
 

data/lib/cloudflare/ai/client.rb

@@ -2,7 +2,7 @@ require "event_stream_parser"
 require "faraday"
 
 class Cloudflare::AI::Client
-  include Cloudflare::AI::Clients::ImageHelpers
+  include Cloudflare::AI::Clients::MediaHelpers
   include Cloudflare::AI::Clients::TextGenerationHelpers
 
   attr_reader :url, :account_id, :api_token
@@ -59,6 +59,17 @@ class Cloudflare::AI::Client
     Cloudflare::AI::Results::TextEmbedding.new(connection.post(url, payload).body)
   end
 
+  def transcribe(source_url: nil, audio: nil, model_name: Cloudflare::AI::Models.automatic_speech_recognition.first)
+    raise ArgumentError, "Must provide either audio_url or audio" if [source_url, audio].compact.size != 1
+
+    audio = download_audio(source_url) if source_url
+
+    url = service_url_for(account_id: account_id, model_name: model_name)
+    response = post_request_with_binary_file(url, audio)
+
+    Cloudflare::AI::Results::AutomaticSpeechRecognition.new(response.body)
+  end
+
   def translate(text:, target_lang:, source_lang: "en", model_name: Cloudflare::AI::Models.translation.first)
     url = service_url_for(account_id: account_id, model_name: model_name)
     payload = {text: text, target_lang: target_lang, source_lang: source_lang}.to_json

data/lib/cloudflare/ai/clients/{image_helpers.rb → media_helpers.rb}

@@ -3,9 +3,18 @@ require "faraday/multipart"
 module Cloudflare
   module AI
     module Clients
-      module ImageHelpers
+      module MediaHelpers
         private
 
+        def download_audio(source_url)
+          download_result = Faraday.new(source_url).get
+          binary_file = Tempfile.new(["cloudflare-ai-automatic-speech-recognition", ".wav"])
+          binary_file.binmode
+          binary_file.write(download_result.body)
+          binary_file.rewind
+          binary_file
+        end
+
         def post_request_with_binary_file(url, file)
           connection.post do |req|
             req.url url

data/lib/cloudflare/ai/models.rb

@@ -4,7 +4,7 @@ class Cloudflare::AI::Models
       %w[@cf/meta/llama-2-7b-chat-fp16 @cf/meta/llama-2-7b-chat-int8 @cf/mistral/mistral-7b-instruct-v0.1 @hf/thebloke/codellama-7b-instruct-awq]
     end
 
-    def speech_recognition
+    def automatic_speech_recognition
       %w[@cf/openai/whisper]
     end
 
@@ -31,7 +31,7 @@ class Cloudflare::AI::Models
     def all
       {
         text_generation: text_generation,
-        speech_recognition: speech_recognition,
+        automatic_speech_recognition: automatic_speech_recognition,
         translation: translation,
         text_classification: text_classification,
         image_classification: image_classification,

data/lib/cloudflare/ai/results/automatic_speech_recognition.rb

@@ -0,0 +1,13 @@
+class Cloudflare::AI::Results::AutomaticSpeechRecognition < Cloudflare::AI::Result
+  def text
+    result&.dig(:text) # nil if no shape
+  end
+
+  def word_count
+    result&.dig(:word_count) # nil if no shape
+  end
+
+  def words
+    result&.dig(:words) # nil if no shape
+  end
+end

data/lib/cloudflare/ai/version.rb

@@ -2,6 +2,6 @@
 
 module Cloudflare
   module AI
-    VERSION = "0.7.0"
+    VERSION = "0.8.0"
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: cloudflare-ai
 version: !ruby/object:Gem::Version
-  version: 0.7.0
+  version: 0.8.0
 platform: ruby
 authors:
 - Ajay Krishnan
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2024-01-25 00:00:00.000000000 Z
+date: 2024-01-26 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activemodel
@@ -107,12 +107,13 @@ files:
 - README.md
 - lib/cloudflare/ai.rb
 - lib/cloudflare/ai/client.rb
-- lib/cloudflare/ai/clients/image_helpers.rb
+- lib/cloudflare/ai/clients/media_helpers.rb
 - lib/cloudflare/ai/clients/text_generation_helpers.rb
 - lib/cloudflare/ai/contextual_logger.rb
 - lib/cloudflare/ai/message.rb
 - lib/cloudflare/ai/models.rb
 - lib/cloudflare/ai/result.rb
+- lib/cloudflare/ai/results/automatic_speech_recognition.rb
 - lib/cloudflare/ai/results/image_classification.rb
 - lib/cloudflare/ai/results/text_classification.rb
 - lib/cloudflare/ai/results/text_embedding.rb